Ladies and gentlemen, welcome to the Roche full year 2020 presentation investor and analysts webinar. My name is Marco, I am the technical operator for today's call. Kindly note that the webinar is being recorded. I would like to inform you that all participants are in listen-only mode during the call. After the presentation, there is a question and answer session planned. You are invited to send in questions for us throughout the entire session using the Q&A functionality of Zoom. In addition to that, you may also raise your virtual hand to address your questions verbally. For participants joining via phone, to raise your hand, use star nine on your phones. Dial that. When you then get selected to ask your questions, please follow the instructions from the phone and press star six to unmute yourself.
At this time, it is my pleasure to introduce you to Karl Mahler, Head of Investor Relations and Group Planning. Karl, the stage is yours.
Yeah. Thank you Marco. Welcome to our full year 2020 analyst call. Personally, I hope that you're all well, safe, and in a good condition. I trust in that. Basically, we do have two hours booked with Roche, one hour presentation and one hour will be Q&A. Most of you are familiar with the Zoom functionalities. Marco already highlighted that. You can raise the hand if you have any kind of questions. You can also use the telephone line. We'll organize this one, and you also can send me an email, and then I can read those things to you. In these unstable times, we wanted to give you some stability, so actually we have the same setup as always, so that is good news. With this one, I want to hand over to Severin. Severin, the floor is yours. Thank you.
Thank you Karl, and welcome everybody to our year-end briefing. Let's go right into the numbers. If I can have the next slide, please. Next slide. Good. You have seen overall results, 1% growth on the sales side, EPS growing over proportionally with 4% and a dividend increase between CHF 0.10. If we move on, please. That's, I feel, an interesting slide, because even though the sides might look a little bit boring at first sight, there was happening a lot below the surface. On the one hand, what you can see is continued good growth on the Pharma side, which actually would have compensated for the biosimilars impact if we wouldn't have suffered from COVID-19, where patients delayed appointments, and as a result of it, there was less prescriptions for medicines. The impact of biosimilars was significant. Some was higher than we originally expected.
Of course, you have seen a very good momentum, in particular in the second half of last year with Diagnostics. I think it's really remarkable that we could increase R&D spend for Pharma with over CHF 800 million. That is an 8% increase versus a sales decline of 2%. It's investments into our pipeline. We have really fantastic opportunities we want to fund, and we could do that with all the savings in other areas, manufacturing, M&P, et cetera. I'd say, yeah, pretty good story in terms of protecting our margins, but at the same time, investing into the future. If we move on. That's a overview of all our contributions over the last year in the fight against COVID-19. All of that happened on top of the ongoing business.
Diagnostics, really fantastic to see the buildup of the portfolio, literally out of nothing, 50 new solutions. You might have seen just this week, we announced a nasal antigen test. We hope that that will be yet another opportunity to broaden testing for more people because it's more convenient, and you don't need a healthcare professional to assist for the testing. Likewise, on the pharma side, important collaborations with Regeneron, with Atea. We'll come back to that in a moment. If we go to the next slide. That's another angle to look at our various efforts in COVID-19. Really the point I'd like to make here is COVID-19 in some form or another will stay with us. We all hope that we will overcome the pandemic, COVID-19 will continue to stay with mankind, like the flu virus.
That's the reason why we are investing, in particular, into the development of a small molecule in this collaboration with Atea. If clinical trial results should read out positively, that would be a very important contribution, potentially during the pandemic, but certainly also beyond. If we go to the next slide, please. Right. Here are the numbers again. You have seen that already, minus 2% on the pharma side, and plus 14% on the diagnostic side. Next slide. The quarterly picture. Next slide. Right. That's an interesting slide because you can see how the business was impacted by the lockdowns. First time in the second quarter with really a significant impact on the pharma portfolio. You see here in orange, the newly launched products where we started off extremely well in the first quarter.
You see the impact where patients just simply didn't go to the doctor or to the hospital anymore, and growth came down. That somehow somewhat recovered in the third quarter. Again, you see a decline in the fourth quarter. Partly this is due because a number of countries were going into lockdown mode again. Partly this is also true to this carryover effect of Ocrevus, which patients are supposed to take every six months. On the diagnostic side, you see the contrary effect because here we have been ramping up our production capacities, and as such, could increasingly meet the demand. If we move on. That's good to see. On the pharma side, an increasing portion of our business is now from the newly launched medicines. On a group level, it's over 40%.
As we see continued good growth with the new products, that should really make us confident for the growth over the coming years. Next slide, please. Profitability. We could keep the margin with all the efficiency gains we had throughout the organization. Next slide. That's again an important one. As we discuss so much about COVID-19 and the pandemic and when it is over and what kind of contributions we make here, we forget that the vast majority of our work is on our other portfolio. Today, still many more people die from other severe diseases other than COVID-19. We keep investing in our portfolio, and 2020 was a really good year in terms of portfolio progress. You see that we transferred a number of new molecular entities into the late stage in our pipeline.
As a result of this, we have now a record of 19 new molecular entities in our late-stage portfolio. That is not counting for the indications. This is just new molecular entities only in late-stage phase III registrational trials. We've never had a stronger pipeline than today, and we intend to move additional molecules into late stage in the current year. That's really I mean, understandably, everybody focused on COVID-19, but I think that's the real story of 2020 when it comes to our mid and longer term prospects. Likewise, on the Diagnostics side, whilst we have been heavily investing into COVID-19, we did not stop working on our new platforms and on our menu of assays. Again, there is a number of very important launches to come, already starting in 2021.
Again, that should provide us with growth for the years to come. Next slide, please. Bill will cover this in much more detail. You have probably seen the more recent data on faricimab, also PDS. Ophthalmology is really increasingly a tremendous opportunity for us. If we move on. In terms of our guidance, in terms of our outlook, good continued momentum with our new medicines in pharma. We expect very good growth in the first half of the year in particular, and then it will depend on how the COVID-19 pandemic develops. That is offset by still a significant impact from the entry of biosimilars. Parallel, we'll work, of course, on our efficiency across the organization. If we move on to the next slide. That gives us the confidence for an acceleration of growth in 2021.
Our guidance is mid-single-digit growth on the sales slide, EPS in line with sales, and that should again allow us to increase dividends in CHF for the next year. Who is next? Bill? Thomas? Bill. Bill, over to you. Thanks.
Thanks, Severin. Thanks, everyone, for joining us. Obviously a very eventful year. I'd like to just underscore a couple of comments that Severin made about the nature of what was 2020. While there was a tremendous flurry of activity regarding COVID, and that hasn't stopped, and we're very involved in a lot of aspects of fighting COVID. However, if you go back three years ago, we were talking about our industry-leading pipeline with 10 exciting molecules in phase III or registration, and now we have 19 molecules in phase III or registration. In fact, we added six in the course of 2020. One of those was related to COVID, but five are in other areas. For example, we doubled our late-stage oncology portfolio in one year in 2020.
I think between the massive shift of resources from operations, from SG&A over into R&D, between that and the progress on actual molecules, I think the real story of 2020 for at least the pharmaceutical division, was one of progress in the pipeline, despite all the obstacles posed by COVID. I'm really proud of the work that everyone's done. We've spoken in the past about our bold ambition to deliver 3x-5x as much patient benefit as we have in the past, but at half the cost to society, and I think we've laid a very strong foundation for that over these last two years, and in particular in 2020, big advances both with the science, the molecules, and the investment profile. We're looking forward to more of that in 2021 and some really exciting results ahead. Let's put up the first slide. Thanks. Yeah.
Again, here you can see the geographic breakdown, -2% overall, and obviously the biggest hits were in the U.S. and Japan, where we had the most biosimilar impact with AH&R all exposed in those markets. Next slide, please. The P&L is actually, I would say, a pretty interesting story to follow, because despite that 2% loss on the top line, we were able to offset that, deliver basically a stable operating profit, but also to invest CHF 800 million incremental in pharmaceutical R&D in the year. I'm really proud of the efforts that we've made over multiple years now, but a lot of those really paid off in a big way.
We did have lower write-offs in the operations area, in manufacturing than we had in 2019, also strong gains on efficiencies, and big gains in terms of our effectiveness and our focus in SG&A as well. Good progress. Next slide. This is what it adds up to. Again, you can see our ambition at the top. We're talking about doubling of medical advances. I explained what we want to do is have twice as many NMEs, twice as many new products, and we want each of those products to have a bigger impact. That's where we get from doubling medical advances to 3x-5x the patient benefit. We're going to do that at half the cost to society.
Again, those costs can be in the price of medicines, but they also can be in terms of early cures, things that avoid future healthcare costs or avoid current healthcare costs. I think we've made some good gains and some good examples of that already. If you look, this requires a lot of internal innovation, and we're investing there, but also a lot of collaboration. We had 92 new agreements in the pharmaceutical realm, five new agreements in late stage alone, and you can see the modalities, and we really have the waterfront covered. You see on the left is small molecules, and we have 12 of those in clinical development across the enterprise. When I say the enterprise, I'm talking about gRED, pRED, Chugai, Spark, and the late-stage organization.
12 small molecules all the way over to personalized mRNA vaccines, two, and a personalized T-cell therapy, one. The grand total of molecules in clinical development is just over 90. I'm super excited about that. I'm, by background, a science nerd, the opportunity to work with over 90 new molecules is a dream come true, we're going to take it up from there. Next slide, please. This is really the revenue picture, obviously the big blue bars at the bottom is the U.S. impact of biosimilars. I think we've seen the worst that it can get. The impact was bigger certainly than our base case, with a total impact in the U.S., Europe, and Japan of about $5.1 billion declines on those three products.
The worldwide impact was about $5.7 billion for the year. We forecast that's probably the worst that it'll get. We think this year in 2021, the number will probably be closer, roughly $4.6 billion. About $1 billion less impact from biosimilars. We do expect to have continued strong growth from products at the top, including Tecentriq, which has continued progress in penetrating a number of indications, especially in the international realm, in Europe, in Japan, outside the U.S. The reason being is because we're already highly penetrated in the U.S. with Tecentriq. Hemlibra, which continues to rapidly penetrate markets around the world, but a lot of growth left there. Ocrevus, we'll come back to, but again, good progress there. I think in this year ahead, yeah, we feel really good about the lineup we have and more to come. Next slide.
This is the Q4 numbers on the left, and you can see we're now up to 47%. It's interesting if you look at the percentages versus the heights of the bar, the bars get squashed a little bit on the right side because of the appreciation in CHF. The bars are actually in actual rates. We see a bit of an impact towards the end, but it's really an impressive performance of these new products, and we added four more new products to the portfolio this year, and we'll either be filing or getting approval for four more in 2021. Again, I think the pipeline investments over many years is continuing to realize benefits for patients and for Roche. Next slide. In oncology, overall minus 10%, really combined impact of biosimilars and COVID. Mostly, in this case, mostly biosimilars as an impact.
I think what I'd point out here is if you look now at Herceptin and Perjeta has surpassed Herceptin. In fact, Perjeta is now our number three medicine. Ocrevus is now number one, and AH&R are numbers two, four, and seven in the lineup. It's quite a big change in a year. Next slide. In hematology, I think this is one to watch. I would just point out here that what's not in shown is Venclexta, and as you know, Venclexta is now a blockbuster. We book approximately half of the profits for Venclexta, but not the revenues. You don't really see that on the bar chart. We've had good progress with our portfolio in hematology, including a couple of things I'll come back to. Next slide. Yeah.
This is some of the latest data from Mosyn and from Glofit, and these are our two anti-CD20/CD3 antibodies. These bispecific antibodies recruit CD3 positive T-cells and take them to the tumor. We've been really encouraged by both the tolerability and the progress on that of these regimens, as well as the clinical benefit. You can see Mosyn in highly refractory patients with follicular lymphoma. This is a breakthrough designation from the FDA. Again, if you look at different patient groups, you can see between 65%- 76% overall response, which is a really great result and a great option for these patients with more indolent disease. In DLBCL on the right, this is Glofit, and I think what was really impressive about this is the dosing is providing for response rates up in the 60s, which you might expect from a CAR- T therapy.
This is an off-the-shelf therapy, again, we're quite optimistic about our prospects for approval of both of these medicines in the next 12- 18 months. We're also, by the way, we're examining a sub-Q dosing for Mosyn that could even further improve the dosing profile. Good progress here. Next slide. Here's Tecentriq. Again, I would remind you that the bar chart is affected by currency because the bars are in CHF, over the last 12 months, there's been about a 6% to 10% increase of the Swiss franc against the major currencies that we're realizing our sales in. You can see in Q4, we had 35% year-over-year growth. Looking forward again, I think a lot of the growth we will see is outside of the U.S. because we have a relatively high level of penetration already in the U.S.
I think the other thing I would mention here is just the high potential for the TIGIT combination because there really hasn't been a big step forward in cancer immunotherapy in terms of the MOA since the advent of the checkpoint inhibitors, and we have a number of phase III studies that could show that type of step forward. That'll have, obviously an effect on tiragolumab, our TIGIT targeting molecule. Because these studies are being run in combination with Tecentriq, then there's the potential for quite a big effect of pulling Tecentriq up with that. On the outlook for Tecentriq, I would highlight a couple of things. The adjuvant studies in non-small cell lung cancer and in squamous cell head and neck cancer. Those studies have been long anticipated. We believe they'll both read out in 2021.
I want to mention that earlier we had talked about the neoadjuvant study for lung cancer, also reading out in 2021. We now think that's more likely to be 2022. The reason being is that study will look at the pathologic response, but also event-free survival. In our ongoing discussions with regulators, we believe that event-free survival is probably going to be seen as more the gold standard. The event-free survival endpoint was always going to be after the pathologic response endpoint, and that result we anticipate in 2022. We do anticipate the adjuvant studies in lung and head and neck cancers to read out in 2021. Next slide, please. Just a brief note, you can see Alecensa continuing good progress. It's very much the standard of care now in ALK-positive lung cancer, and we're really pleased with the continued growth of Alecensa.
You can see in Q4, we were up over CHF 300 million, so well into the blockbuster category, again, really pleased with the durable benefit of Alecensa for these patients. Next slide. Hemlibra, I think this is a really good illustration of the effect of the pandemic. You saw very strong growth really across countries up through Q1 of 2020. In Q2, basically, almost no new patients going on and some existing patients having disruptions. Q3, actually a very strong rebound, Q4, again, healthy growth. We have every expectation that Hemlibra will continue to have strong growth as it penetrates in the inhibitor markets in the international area, but increasingly in Europe and the U.S., penetrating outside inhibitor. Look for continued growth here. Next slide. All right. Immunology.
What was a little surprising on immunology in 2020 is that we weren't surprised by the reduction in Rituxan sales due to biosimilars, but obviously Actemra upside there based on the use of Actemra in many countries for treating critical patients with COVID, and those studies continue to read out. I'll say a little more about that in a bit, but you can see the overall immunology franchise at plus 2% in constant exchange rates. Next slide, please. Ocrevus. Basically, here's what's going on in the MS market from what we can see, is that people with MS, especially relapsing MS, are dealing with the disease for 30- 40 years. There was a high level of switching to Ocrevus up through Q1 of 2020.
We were up in the high 30s to 40s in terms of our percentages of new and switched patients, and there were a lot of those patients switching. Beginning with late March of last year, the switching slowed dramatically, and in addition to that, there were a lot of existing Ocrevus patients who were nervous about whether or not to get a dose who delayed their doses because of either concerns about safety or concerns about the therapy itself or concerns about going into a hospital or infusion center. In Q3 of 2020, we saw those returning patients who delayed their doses in Q2, they all came back in Q3. We got a bounce in Q3, and you can see that reflected here.
In Q4, essentially what happened was that the patients whose doses got delayed out of Q2, that meant they were also delayed out of Q4, right? If you were supposed to get dosed in May and you didn't get dosed in May, but instead on July, then you also didn't get dosed again in 2020 at all. Your next dose shows up in January of 2021. In addition, the rate of switching is still lower than pre-pandemic for sure. It's hard to estimate exactly how low, but when we've done our market share survey, and it's a monthly market share survey, we have the data through October, and we follow a three-month moving average, and that number was about 40% for October.
We don't see any reduction in or significant impact on people's interest in Ocrevus or their choosing of Ocrevus versus the other 17 alternatives in the market. It still remains very high, the rate of switching is still lower, I think as people are kind of riding out the pandemic and waiting to see what to do next. Overall, we have strong confidence in the future of Ocrevus, and we expect that we'll have good growth in 2021, and that we'll continue to see this cycle because that actually could last for years where we'll have every other quarter up and down. Next slide, please. All right. Evrysdi. This has been very well received. There's no other way to put it.
We have more than 350 doctors in the U.S. that are prescribing Evrysdi, which is quite remarkable because there's not so many doctors who treat SMA. Virtually all the specialty centers are now using Evrysdi broadly. We've got about 1,000 patients treated, that was after less than five months in the market. This is a launch in the middle of a pandemic, which I probably don't have to tell you is a bit more of a challenging exercise. Frankly, the reception has been overwhelming. We're getting quite a few naive patients. About 1/3 of the patients are new to therapy. Those tend to be older patients, children or adults. About 2/3 though of the patients are previously treated with one of the existing therapies. Again, it's a broad range of ages from two months old up to 70+ years. About half the patients are adults.
Again, we think this is going to be a very important therapy for a long time in the future. It's very well tolerated. It's a convenient once-a-day oral dosage, and it doesn't require contact with a healthcare provider. Again, it's been a really good start. We'll have more data coming this year in the previously treated patients. We'll have the two-year data from the pivotal studies, and we have ongoing study in newborns. Really excited what we see. Next slide. I should mention Evrysdi, we are still looking forward to approval in EMA in the first half of this year and in many other countries around the world. Let me make a few comments about ophthalmology.
You've noticed since our last quarterly call that we've actually had four phase III studies with faricimab have read out, two in DME and two in AMD, and we were, again, very pleased with what we're seeing. About 50% of the patients with DME were able to achieve 16-week dosing. About 45% of patients with AMD achieving 16-week dosing. This has really never been seen before in trials of molecules targeting angiogenesis in retina. It's a CHF 12 billion global market, and we think that faricimab is going to be a really important new choice for physicians and patients. Some people might ask, "Well, how does that relate to the port delivery system?" Again, in May, we showed data that we had about 98% of patients getting six-month dosing intervals with the port delivery system and how do these things all fit?
I think one way to think about it would be faricimab is a very well understood route of administration, right? It can be given the same way that Lucentis and other therapies are given in a simple dose, but with a 16-week interval. The Port Delivery System, that'll be really great for the half of patients that are able to get a 16-week dosing interval. Other patients may need a 12-week, eight-week, even four-week dosing interval on faricimab. Those patients would be great candidates for the Port Delivery System. They have the implant done, they can have basically twice-a-year dosing. We’ve also announced a study of the Port Delivery System on a 36-week dosing, which would extend the time out to nine months for the PDS.
We just think that more choices are going to be really important. This is a really large therapy area, and there's a high unmet need for continued efficacy over time. Next slide. Let me just touch on the infectious disease area, and obviously, this is something that's on everybody's mind these days. I'll start with Actemra maybe because that was the first thing that we studied. As you know, we've run a number of randomized controlled studies. We've had mixed results, and I would say we think a good part of that has been due to different endpoints and different patient populations. We think we're sort of zooming in on both the most relevant endpoints and relevant patient population. It seems like the ideal candidates are patients who are really in that acute phase of inflammatory attack.
They're characterized by needing high-flow oxygen or some ventilation support, but not yet requiring organ support, not too early. The REMAP-CAP study read out in January. That was announced in England. We don't have the full data on that. That's actually a study being run by the NHS. Based on that study and what they reported was a 10-day lower time of hospitalization, also lower need for mechanical ventilation and other important endpoints. Based on that, they've authorized Actemra on the NHS, which I'd say is a fairly high bar. We have two more phase III studies reading out for Actemra and COVID over the next six weeks or so. Those should really confirm what we have.
The Regeneron cocktail with casirivimab and imdevimab, and you may have seen the news, but there was some studies published in the last week that looked at the ability of these molecules to overcome the resistant variants. So far, at least the cocktail is holding up. I think that validates their decision and ours in working with them to really focus on having two or more molecules to target COVID so that we can overcome resistant variants in the future. Next slide. In terms of the final say on 2020, we were able to add a couple more green checks. We were pleased on the key outputs, including those two phase III studies for faricimab. It was a year where we had some big wins and then some challenges and some setbacks.
We know that if we're not taking big risks, then we're not going to drive medicine forward, and that's really what it's all about for Roche. I think you know that we have a real passion for innovation in medicine, and some years we're going to have some more red Xs. We have a lot of excitement about what's in the pipeline and our prospects. I'd say we ended the year in a reasonable place. If you go to the next slide. This is some of what's coming up, and I think, again, the additional studies that are going to read out on outpatient and prophylaxis of the cocktail. That's significant. I mentioned already the adjuvant studies of Tecentriq. I didn't talk about POLARIX, which is Polivy plus R-CHOP in DLBCL. This is a big deal.
It has been 20 years since Rituxan was established as the standard of care in frontline diffuse large B-cell lymphoma, and now we have an opportunity to best Rituxan. That's what we want to do with Polivy. Looking forward to that readout mid-year. Also Mozy and Glofit, some important data readouts there that will enable filings. Continued, as I mentioned, the switching data on Evrysdi. I think quite a good year for news flow, and we think it is going to leave us in even stronger position at the end of the year than we are starting. With that, I think I want to turn it over to Thomas.
Thank you, Bill. Hello, also from my side, I hope that you had a good start to the year despite this terrible pandemic and all the personal consequences that everyone has to deal with due to this pandemic. If we go to the next slide. I'm happy to present the full year Diagnostics division performance. With sales of CHF 13.8 billion, we had more than 14% growth for the year. This growth was driven predominantly by molecular diagnostics with 90% growth and point of care, which is part of the Centralized and Point-of-Care business area, which grew 212%. This was driven by antigen testing, really, we only had the antigen test available the last two months of last year. On that regard, having 212% growth for the year is really significant. Severin already showed the growth by quarters.
Q4, we had more than 28% growth, and we do expect further acceleration of this growth, particularly in Q1 and Q2 of this year. Now, centralized and point of care declined by 1%, and this was due to a decline in the routine testing due to the pandemic and the lockdowns. If we go back during the quarters, in Q2, we had minus 17 in this business, minus 17%. This was really the hard lockdowns all over the world with 30%- 40% volume drop in the months of April and May. June was much better. You can see how well this business recovered, but obviously also the antigen testing did play a role.
By the way, we already saw a Q1 impact at the time because in China we had the first impact, and here we had specifically in March, 2020 a significant impact in volumes in this business. Sales in tissue diagnostics grew 5%. Here we had fantastic Q1, Q3, and Q4. Q2 was really impacted by the lockdowns. Otherwise, in the other quarters, much higher growth even. This business doing extremely well. Diabetes care sales declined partly due to COVID, but also because of a continued adoption of competing technologies to BGM. Overall, we do expect the COVID-19 portfolio to grow significantly also in 2021, and particularly in the first half year.
We also expect that the non-COVID business, the kind of core business, will grow very well in 2021 because we did have certain impacts this year, and we will get a certain tailwind out of that. Now going to the next slide. Here, this is the regional split. Sales growth was driven strongly by North America, EMEA, and Latin America. In these regions, we did have a lot of COVID testing sales business. At the same time, we did have an impact in our routine business, but net, it was definitely positive. We did see a recovery of that routine business, particularly in Q3 and Q4 over the time. Asia Pacific is the only region that actually had a decline, again, due to these preventative measures, particularly in China, which was at -11% for the year.
China was the first impacted and really didn't have much of the upside in terms of COVID testing sales beyond the MagNA Pure and LightCyclers. If I take China out, actually, APAC also grew double digits. One other specific information on China is that we actually did also take the opportunity to further lower inventories from 80 days inventories to 45 days inventories, which is approximately CHF 200 million, and most of that was happening in Q4. With 45 days, we're at the lower end of what we need in terms of inventories in China. Going to the next slide. Here, just focusing two areas, centralized and point of care, which decreased by 1%, mainly due to immunodiagnostics and clinical chemistry. Again, largely impacted by China and also by this inventory reduction.
If you look at point-of-care immunodiagnostics, two months of sales in antigen and for the entire year with that we grew 667%. Really strong impact here. Molecular grew 90%, mainly due to the PCR testing. Within the virology line, we report the tests for the high throughput, 6800, 8800, which in Q4 grew 250%. The Lite Mix systems, this is where we report on the MagNA Pure and LightCycler and this grew by almost 200% in Q4. Going to the next slide on operating profit. Here we had substantial growth in operating profit by 50% and thereby we grew the operating profit faster than sales. This was on one hand, a favorable product mix. Also as you can see, we had very good and tight cost management across the organization. Cost of sales you see with the 10%.
This on the one hand is of course higher volumes, but also, we sold more instruments than we would normally do, and the margins on instruments are lower. The third piece is really higher costs in terms of global supply chain. To really bring this product faster to different places, also flights and containers were more expensive than they normally are, and we really needed to bring the products where they were needed as quickly as possible. M&D is flat, but here again, if you look at marketing and sales, was actually down. Here also the local distribution costs was rising significantly. Again, superb cost control. R&D, this is what we want to do. We want to invest in new products and this was really driven on one hand by all the COVID products that we launched, but also, we didn't neglect the rest of our portfolio.
We really pushed forward on that. You will see 2021 will be a fantastic launch year. This is really paying off. G&A is slightly above zero, and this is due to legal expenses in the U.S. Also here, really going in the right direction. Overall, very happy with this development. Now going to the next slide. Here, you can see our overall portfolio on what we have launched. The virus was not even sequenced, and the sequence was not known in the beginning of the year. It was only the very couple of first weeks of January when this sequence was published. To develop this many products in such a short period of time and at that quality was really a huge effort by our organization working day and night.
The ramp up in production, I have to say, the team really did everything they could to help the world fighting this pandemic. We have developed those solutions on the molecular side, so PCR, and on the immunology side, which is antibody and antigen testing. Antigen to detect the disease, antibody to detect previous infections, but also in clinical labs and point of care. Really a comprehensive solution for everyone, and they can pick the options that they need in order to fight the pandemic the best way in their country. Let me highlight just three things on the slide. Two of them, I'll get back to it on a later slide. One is the anti-SARS-CoV-2 S antibody test, which got emergency use approval in the U.S. This is very much needed, specifically in conjunction with vaccines that also target the spike protein.
The second piece is the authorization for the Elecsys antigen test. Being able to detect the antigen on these lab machines that are basically present in every hospital around the world, very quickly, very precise. Again, another weapon to diagnose people quickly, specifically healthcare workers, et cetera. A very important one. Just beginning of the week, we also announced that we received CE mark for our rapid nasal antigen test. Again, very important weapon, specifically because it allows for much easier collection of the sample. Obviously with that, it's more convenient and people will be more willing to do it on a more regular basis. This is a great opportunity to then open up more of society and more of the industry as well, because you can control infections much better. Going to the next slide.
This is something that we didn't talk about that much last year. We always said that we were working on a ramp up, but we wanted to take this opportunity now because now it's happening. Normally it takes 18 months to really do this. We took a risk back in March, April, invested more than CHF 600 million, hired more than 1,200 people, and really focused on, first of all, buying 90 manufacturing lines from our suppliers, then building buildings, and validating those lines, and the first of those lines now came in end of last year and the beginning of this year. What we could do in the beginning of this pandemic is really maximize the usage of our existing lines, which means people worked day and night, not only in our organization, but also with our partners. People were not taking vacation.
They were working weekends, they were working during holidays between the Christmas and New Year. There was only one day where people didn't work. Really, they are giving everything to make sure that we can do the best for the world, and they're doing it with pride, and they're doing it out of free will because they know it's important right now. This was really a huge effort. When you talk about the test, often people think, "Oh, it's only one test that you have to ramp up." That is not correct. We actually need 17 different products that we need to ramp up. There are multiple different consumable pipette tips, different plates, et cetera. There are raw materials that you have to ramp up. Yeah. There is a lot that you have to do.
In a typical test on its own, only the test itself, there are 600 components. This is really complicated. To do this in such a quick time was extremely important. Again, these are very specialized clean room, automated manufacturing processes, and this would normally take much longer. What would that mean now? Our wrap up will actually result in a doubling of our PCR capacity by end of Q1, and then again, a doubling approximately by half year. I just want to say that we obviously did the mathematical theoretical calculation, what that would mean in sales, and these are obviously mind-boggling numbers. There are other factors that make it a bit unclear on how things are going to develop.
We did not include all of the sales in our outlook because we know that testing sales can be impacted by the rollout of vaccines, can be also impacted by the different variants and mutations that are coming, and can be impacted by pricing. There are things why I think we're well-positioned, one of them around pricing. We are charging less than the average other company out there, so we have a certain buffer in there that others will have to go down significantly to match the price. That's one end. Also, a lot of the testing that's being done as well is very manual, and with that, also very cost-intensive. The quality, if it's manual, is then also not at the same level when you have such high level of automation that we have with our systems.
With our systems, you can take the sample, put it on the machine, you can go home. You don't even have to do any other manual intervention. I believe as potentially, volumes will go down, people will switch mostly to our systems because it's just a lot easier. There are certain things where I see as a potential, but from our outlook, we really, mostly included sales of COVID testing in the first half year, and then Q3 and Q4 would then be a potential upside. Going into the next slide. Some of the portfolio that I wanted to highlight around COVID. One is the Elecsys SARS-CoV-2 antigen test. Here, this is a new solution. It's an immunoassay for the qualitative detection of the nucleocapsid antigen of SARS-CoV-2.
As you can see, the performance is truly excellent with almost 3,000 patients in our clinical trials. What I think is also special about it is the virus inactivation time. Right now, there are not many players that have such a solution on the market, and most have a virus inactivation time that's more than an hour. We have a virus inactivation time of two minutes. This solution can work in all the hospitals. They can screen healthcare workers on a regular basis, if they see the virus, and the results are available in 18 minutes, and you can do up to 300 tests an hour. That is really significant and another important contribution to fight this pandemic. Next slide, please. Here, we launched in December our Elecsys SARS-CoV-2 anti-S antibody test, which received emergency use authorization.
This antibody, compared to the nucleocapsid one that we launched in May, targets the receptor binding domain of the spike protein. It's anti-S. Our nucleocapsid assay was already used, for example, in Moderna trials to baseline the start of the trial. They used that. Now they're also using this assay to monitor the level of antibodies and also the duration on how long this vaccine is going to work. We're working very much with these vaccine companies on this. You can all see the results on the performance, which is confirmed also by these companies, is really excellent. We developed this assay before there was an international WHO standard in terms of quantification, because this is a fully quantitative assay. The good thing is, WHO came out with a standard last month, it's a perfect correlation to our assay.
This is really good news, because everyone that will develop something like this will have to standardize against this international standard. Moving outside of COVID, into oncology, we recently also launched a PIK3CA mutation test, to enable fast decision-making for targeted treatment decisions. In advanced or metastatic breast cancer, PIK3CA mutations are often associated with tumor growth and resistance to endocrine treatments. This test detects 17 different mutations in the PIK3CA gene and can also help identify then patients who can benefit from a PI3 kinase inhibitor-targeted therapy. As you may know, PIK3CA is one of the most commonly mutated genes in advanced or metastatic breast cancer. Up to 40% of the patients do have mutations in this gene. Our studies have shown higher analytical sensitivity and clinical reproducibility. Going to next slide.
With nearly 2.1 million cases of breast cancer that are diagnosed every year and 15%-20% of those being HER2 positive, it's obviously important that people get very good and fast diagnosis. If you look at those slides, there may be some very good pathologists that can read that very accurately, but clearly with the rising amount of cases, it's important that we digitalize this and standardize this. With these algorithms that use the latest advances in artificial intelligence, we can do that. We have launched three this year. We're going to launch more next year. This can be used in conjunction with our scanner, the DP 200, but also is run on the uPath software and the information can be transmitted into the NAVIFY Tumor Board. Going to the next slide, giving you a bit more of an outlook.
I believe 2021 will be an amazing year in terms of launches for us. Here you have three system launches. We, in addition, will also launch another system called the Cobas Pulse. Beyond even the system launches, we have really a fantastic pipeline of medically valued assays and solutions that we're going to launch. We did this exercise looking at how much we launched, not in 2020, because we launched more in 2020, but the years before. We launched an average five of these medical value assays. 2021 is going to be 17. This is really fantastic to see this progress. What I also think is labs need great systems that are easy to use to be able to run those. On the Serum Work Area side, we had the cobas pro, which was the next generation system.
Now we're launching a system for the high throughput with the cobas pro ultra-high throughput and the cobas pure for the lower end. This is unique because many of our peers, they play maybe in one of the segments mostly, but to really have a family approach across all is really what our customers need because they think in networks. The same we did on the molecular side. We had the cobas 6800 and the cobas 8800, which have been proven to be perfect for a situation of a pandemic. I would say there's no peer out there that has systems in that throughput range. Most were playing more in the area where we're going to play on the cobas 5800. We had, I would say, somewhat of a gap here because we had, with the cobas 4800, not such an automated system.
To now launch this, we have, again, a full family approach and really something that is extremely competitive in this space. Again, all of these systems have identical user experience, identical reaction concept, same performance, same quality, allowing standardization across. Really excited about this. Next slide, please. Here, I would really like to invite you to our Diagnostics Investor Day, which is going to be on the 23rd of March, and will be taking place virtually in our new customer experience building in Mannheim. You will get the feeling that you're actually in that building. In 2020, we did launch our new diagnostic strategy, and with my leadership team, I would like to give you the opportunity to see that, but also give you an look into really our exciting pipeline. Yeah, I can just invite you to that. Next slide, please.
The team did a tremendous effort to not only deliver the 15 COVID solutions, we also were very productive in the rest of our pipeline. This will definitely ensure a continued success of our business. Last slide, please. I'm also very excited on how 2021 is going to look like. You can see that the font is getting smaller, and that's because we have a lot of launches. I believe 2021 will be huge, and also this will drive our growth in the following years. Really looking forward to your questions later, and I hand over to Alan.
Thanks, Thomas. Great to see the excitement and great to see how diagnostics is flourishing and what you've really done, in case of contributions, really to testing and to society, I would even argue. Really great. Welcome from my side. I hope everybody is safe and healthy. I think you can imagine I'm excited about the contributions that Roche made in 2020, and certainly about all the contributions to come, really to create more safety and health in our society. With that, let's go to slide 55. You've seen the agenda very quickly. Nothing unusual. The highlights, I will tap on all of them. Let's move on to the next slide. Here you see really what Severin has alluded to already, I think that the major shuffle, reshuffle, if you like, of resources in our company.
I think on the sales side, as you mentioned, a minus CHF 1.2 billion sales reduction on the Pharma side. Nevertheless, I think extreme growth from the new products, you've seen that. Certainly the Diagnostics division was CHF 1.8 billion up, CHF 2.6 billion of COVID-related sales, which I think also shows that there was really an impact on the underlying business on the routine tests. On the right-hand side, you see the profits. Really, I think what is amazing is, on one hand, I think the momentum that you're seeing and the profit growth. The other piece is really here that Pharma lost the sales, that I've mentioned already, invested CHF 803 million more into R&D, and then really balanced that out with Pharma efficiencies and other gains in that area to stay basically flat on the Core operating profit.
You see really Diagnostics coming in with a significant growth in Core operating profit of CHF 981 million, which really resulted in the 4% growth as mentioned. Good. With that, let's go to the next slide. This is the comprehensive view. Let me very quickly lead you through this. I think sales in constant rates are up 1%. I think my colleagues did a great deal in explaining that already. The core operating profit up by 4 percentage points. Good cost management, I've explained that a little more into detail. The core net income by even a higher dynamic, with +5%. Here the financial result plays a role. We'll dig into that as well. You see really the core EPS loses a little bit of dynamic in constant rates by up +4%.
Here the point is that certainly you take the minorities out for Chugai, and I think we all know that Chugai had a great year in 2020. The IFRS net income up 17% in constant rates. This is certainly due to a base effect coming from 2019, where we wrote off the goodwill for the diabetes care business. You will see that later on. Operating free cash flow down by 21% in constant rates. I will dig into that. In my opinion, not a major worry, because very well explainable. You will see an increase on the net working capital side, which I think really we can benefit from in the outer years when that really turns into cash. With that, let's go to the sales bridge here. Nothing, I think unusual.
You see really Pharma now in a little bit more detail from a regional point of view. You see the impact in the United States, certainly driven by the biosimilar impact. You see Europe, which could overcompensate the biosimilar impact. International with solid growth, I would argue, especially in China. Chugai also impacted by biosimilar competition, and then the diagnostics growth that I've mentioned. You see really the group growth of CHF 620 million. You see the significant currency impact of CHF 3.8 billion, which I will allude to later on as well, which gives us a -5% in CHF. Good. With that, let's go to the core EPS development, the bridge over here. Let me start really with the point that I think overall, core EPS has grown by 4.2%, from CHF 20.35 to CHF 21.20.
You know that when you look at half year, I think we had a momentum of 1.9%. We even increased the momentum in the second half. Let me lead you through this. I think the first quarter is about income from disposal of products. We had less disposals compared to last year by roughly CHF 250 million. At half year, we had a larger negative impact here of roughly CHF 300 million. We did some smaller things in the second half, but still, let's say a gap of CHF 250 million to last year. You see other royalties and operating income that's basically flat. Same applies to the gains in equity securities, completely different composition, but basically flat. Then the bond redemption that we did in December of 2019 had an impact of -CHF 202 million.
You see really that gave us a base effect in the second half when it comes to core EPS. Resolution on tax disputes, I have a slide on that. I think we had positive impacts from resolution of tax disputes in 2020 as well, like in 2019, but a little bit lower. A little bit lower, admittedly, that took a little bit of momentum away. Look at really what we contributed on the operation side to bring the core EPS up, an increase of + 4.9 percentage points. Good. With that, let's go to the P&L. Let's jump right away to the royalties and other operating income. I gave the explanation already, really lower income from the disposal of older products. The CHF 250 million, I think that explains that line. You see the cost of sales.
On the cost of sales, really, I think you see a reduction here of CHF 1 billion, so a saving of CHF 1 billion. There is an impact from the Pharma side of roughly CHF 1.6 billion. On the Dia side, an increase of CHF 600 million. That really nets to a saving of CHF 1 billion. I will explain that because that deserves a little bit of explanation on one of the next slides. M&D in Pharma down by CHF 600 million, flat when it comes to diagnostics. R&D, an increase of roughly CHF 1 billion, CHF 953 million, CHF 803 million on the Pharma side and the rest on diagnostics, which I think is amazing because you really see how we reinvested the money in a time where we faced a lot of challenges. You see G&A, and that is, how should I say, that could be worrying.
It isn't, because the CHF 157 million is driven by Spark. We consolidated Spark relatively late in the year 2019, in the second half of December. Basically no effect here. In 2020, we had the full base effect of Spark, which really explains CHF 100 million of that increase of CHF 157 million, and the rest is really in FMI and Flatiron as they build their organizations. The Core operating profit, I've mentioned a couple of times already, up by CHF 981 million, which represents an increase of 4%. Let's dig a little bit into things. I think on royalties and other operating income, I can be quick. You see royalty income, out-licensing income, and other operating income is balancing out, as you can see. The other operating income is driven by Venclexta. I think Bill made a comment about that product.
I think that product is becoming really meaningful when it comes to profits and has contributed significantly on the profit line for Roche and for pharma. Then you see really the income from the disposal of products, which I've mentioned already, which was by roughly CHF 250 million lower compared to last year. With that, let's go to the cost of sales. As said, I think that deserves a little bit of explanation because you see overall, and I've mentioned that in the last line here, in bold group, you see the saving of CHF 1,015 million. When you go to the pharma division, you see a saving in constant rates even of CHF 1.6 billion, CHF 1.7 billion, if you round precisely. Certainly you actually say, "Wow, that's a big figure." Let me explain that a bit.
I think really first of all, you see the manufacturing cost went down by CHF 800 million, roughly. That's the plus CHF 787 million. I think we had a couple of base effects here. We had great efficiencies. I think this is really what matters here. The rest, so over CHF 800 million, basically half of the overall effect, is really driven by the fact that we paid lower collaboration and profit sharing. I think that was one element, which came down. That is around CHF 400 million. The other point is lower royalty expenses. Well, Cabilly came down or Ocrevus came up a little bit with the royalty expenses, which is understandable. I think overall that gave us a saving of around CHF 200 million here. I think just to put that into perspective. As Bill said, I think a major improvements in efficiencies.
In Dia, it's a pretty straight increase of CHF 637 million, which is also represented in the manufacturing costs. With that, let's go to the core operating profit and the margins. You see really the margins went up in all divisions, not just in constant rates, also in CHF. Let me point here on the pharma side. I think that's amazing because, as said, with lower sales and a higher investment into R&D of CHF 803 million and still defending the margin, I think is quite an achievement. Look at diagnostics and look at the group overall. With that, let's go to the core net financial results, and the core net financial result has improved quite significantly, as you can see, by CHF 339 million. Let me lead you through the bridge. You see the net interest income has deteriorated.
Well, we had lower cash on hand in 2020, and interest rates went lower. Equity securities basically flat. Currency, no impact. You see where the debt redemption, as mentioned already, which we didn't do again in 2020, but did it in 2019, so a base effect in 2020. You see where the interest expenses, which is certainly to a major portion, also the result of the debt redemption that we have done. An improvement of CHF 182 million, and then you see the position of that, which is a nice mixture of a lot of things. With that, let's go to the group tax rate. The group tax rate in the middle, you see that when you look really at the underlying group core tax rate, not a lot of move from 18.4% to 18.6%.
What you also see in green is that in both years, we had quite some help from positive impacts from the resolution of tax disputes. You see really in 2019, we came out with 16.3%. The resolution of tax disputes meant a reduction of the tax rate of + 2.1 percentage points, representing an absolute amount of CHF 454 million. You also see in 2020, where the reported rate is 17.1%, that we had some help here from the resolution of tax disputes coming in here by a reduction of - 1.5 percentage points, representing a positive impact of CHF 317 million. When I look at 2021, let me say here, I expect a group core tax rate of about 19%. See a little bit of an increase in Japan, in Chugai, that plays into this. Roughly 19% should be in the cards.
With that, let's go to slide 67. This is about the non-core items. You see the core operating profit increase, which I've mentioned, plus 4% in constant rates. You see the global restructuring plan, basically comparable to what we have seen in the past years. Amortization of intangible assets, a little bit up, driven by ROZLYTREK, which came new to the portfolio, if you like, here. The impairment of intangible assets, which went down significantly by over CHF 1 billion. Here the major driver is that we have written down the diabetes care goodwill in 2019. M&A and alliance transactions flat, basically. Legal and environmental with quite some swing of CHF 800 million. You see the negative impact there in 2019 of minus CHF 480 million, very much driven by one legal case we reserved for Meso on the Dia side.
I think really in 2020, we released the Accutane provision of roughly CHF 300 million. CHF 347 million here positive. I think that all adds up to quite a positive year of roughly CHF 2 billion from the non-core item, which brings the IFRS operating profit up by roughly CHF 1 billion, +16% in constant rates, which is reflected in the IFRS net income, which goes up by 17% in constant rates. Good. With that, let's go to cash on slide 69. As said, I think a significant impact from CHF 20.9 billion in 2019, the operating free cash flow went to CHF 14.8 billion in 2020, in Swiss francs. You see the first point to make is foreign exchange of CHF 1.7 billion. That leaves us with two major impacts.
One is investments into intangible assets. I think that's a good thing because this is what supports our pipeline, especially on the Pharma side. When you think about Sarepta, Blueprint Medicines, Atea, I think this is all investments which are reflected here. If you see really the increase in networking capital, the major driver here is inventories up by CHF 1.4 billion in both divisions, in Pharma and basically equally in Diagnostics. On the Pharma side, it is really about the new launches. On the Diagnostics side is that we have good inventories for all the COVID tests and what we're doing over there. I think that's justified. No doubt that this will turn into cash over time. Another piece here I should mention is that we have taken lower provisions in 2020. With that, let's go to the group net debt level.
I think, well, we are not net cash positive. We still have net debt on hand with a minus CHF 1.9 billion. Start of the year with a minus CHF 2.5 billion. You see the bridge here. I'm not going through all of that, because I think that's pretty obvious. Let me make a point about the bar on the lower part of the slide. Because what we have seen in 2020 is that we had quite of a trade-off between intangible assets, investments into intangible assets and M&A. Perhaps that was a reflection of all the high valuation in the market, that we did more in-licensing deals, and agreed on more milestones and more sharing of opportunities and risks, moving forward.
You see really, when you look in total, at our investments into innovation, there is not a huge difference to what we have invested, at least from a balance sheet point of view in 2019 compared to 2020. Good. With that, quick look at the balance sheet. Balance sheet, nothing really extraordinary. Cash and marketable securities pretty flat. Other current assets, up by CHF 1 billion, roughly. That is basically inventories. The non-current assets, this is the increase in intangible assets of more than CHF 1 billion. Then you look at the liabilities. The current liabilities, went up by short-term debt of +CHF 2.3 billion. The non-current liabilities went down by long-term debt by -CHF 2.4 billion.
You see really the equity portion now at 46% and roughly CHF 40 billion, when you look at the absolute number, which I think is a nice improvement compared to how equity looked like when I joined the company 10 years ago. Net debt to total assets at 2%. Good. With that, quickly to currency on 73. When you look at currency, well, we've taken quite a hit in 2020. Let me emphasize once again that we have a pretty good natural hedge, which means we have our major sales in currencies where we also have our major costs. When you look at the U.S., we have a full supply chain, a full value chain. In the U.S., we have a full value chain. In China, we have a full value chain, when it comes to Europe.
I think that's, for me at least, a little bit more of a reporting point. When you look really at the impact in 2020, and you see that on the right-hand side with a minus 6% on sale, 6 percentage points on sales, with a minus 8 percentage points on operating profit and a minus 9% on core EPS. If we assume that all currency rates stay flat, from year-end 2020, we would expect impacts between 3 to 5 percentage points, on sales, operating profit, and core EPS in 2021. Good. Core EPS, slide 74. Let me set the stage for your assumptions and for the analyst assumptions for core EPS in 2021. To do that, I think we have to get the basis right for 2020. Core EPS 2020 as reported is CHF 19.16.
You see now that you need to take an adjustment for the foreign exchange losses, as we apply and all apply to a constant exchange concept, exchange rate concept, to be precise. To get to this CHF 0.919, to eliminate that, you take the currency losses of minus CHF 206 million. You find them on page 65 of the finance report, and you take the taxes away from them. That's roughly CHF 38.31 million. You get to that by when you multiply the CHF 206, the minus CHF 206 to be precise, with the underlying tax rate of 18.6%. When you do so, you get to a CHF 167.68 million, and you divide that by 865 million shares. That means voting and non-voting shares. You will find that number on page 171 in the finance report. Once you do that, you get to CHF 0.194.
That is exactly the number that we have highlighted here in the slide. You have to add that up to get to the CHF 19.35, which then represents the basis for the projections for 2021. Good. With that, I think the last slide is the outlook. I think Severin has alluded to that. Looks a bit boring, but I think given all the uncertainties and the momentum and all the challenges we have seen in 2021, very happy to provide that guidance. I think we all look forward to your questions now. Thanks.
Yeah thank you. Could we get again, the administrative illustration here on how we can ask questions and participate in the call? Operator, please.
Yes, of course. You are invited to send in questions for us throughout the entire session using the Q&A functionality of Zoom. In addition to that, you may also raise your virtual hands to address your questions verbally. For participants joining by phone, to raise your hand, use star nine on your phone's dial pad. When you then get selected to ask your questions, please follow the instructions from the phone and press star six to unmute yourself. With that, back to you, Karl.
Yeah. Thanks a lot Marco. Just for the record, we had at peak about 870 people joining. It's still well above 800- 830. That is really an excellent participation. I just wanted to already thank you for your interest in Roche. Maybe let me get through the questions. If you could kindly maybe limit your questions to two. We will try to keep our answers short so that we make best use of the remaining time and get through all your questions. The first one I got via the web here, and also via the chat, from Michael Leuchten and from UBS's Sam Fazeli. They had basically the same question, so I just try to summarize it, and it's going to you, Bill. They were wondering about the impact on Avastin, Herceptin, and Rituxan during the year, but in particular in Q4.
If you could maybe illustrate a bit what happened in Q4. Maybe if you could give it a try, what happened to COVID? What happened to foreign exchange? What happened really to the biosimilar? That is basically the line of questions which we got from these two gentlemen.
Sure thanks Karl. I'm not sure there's too much color I can add to it. Essentially, the impact that we saw in the U.S., which was obviously the biggest country impacted last year, was, I would say, rather linear over time. As the year evolved, we weren't sure. We thought, "Oh, maybe it's leveling off a little," and then it would accelerate again, and then it would level off a little, and then it would accelerate again. If you step back and look at the whole year, it was pretty linear. The reason Q4 was such a large impact is just the impact in 2019 was rather small, and the monthly impact was just growing over the course of the year following down that curve. Q4 just had a big number. I don't know that there was anything really special there to point out.
Okay.
I saw another question that was sort of related, maybe I answer, which is just why was the number bigger than we had projected? I think we started the year saying that we thought the number for U.S., Japan, and Europe would be about CHF 4 billion, and it ended up being about CHF 5 billion for those territories. Honestly, I think the reason that it was bigger than we expected was that we thought that maybe the U.S., and I think this was very consistent with what we said, we always said the U.S. would be like Europe, but maybe a little less steep. In fact, it ended up being pretty much like Europe.
We gave the erosion a little bit of a haircut and thinking that maybe in the U.S. it would be a little slower, largely just because of systemic factors in the U.S. healthcare system and some of the incentives and things. As you're aware, there were a lot of other biosimilars that had very poor uptake in the U.S. We always said that you shouldn't expect that with AH&R because some of the reasons those other biosimilars weren't used were kind of historic reasons. In the end, yeah, the uptake was quite large, consistent with what we saw in Europe. I think it's a new chapter, but fortunate for us, we've got a strong pipeline and able to continue growing those new products right through.
Yeah. Thank you. I wanted to take a call here via the telephone line, Matthew Weston. Matthew, I open your line now. Matthew?
Can you hear me, Karl?
Yeah. We can hear you. Yeah.
Perfect. Thank you. I've got two questions, please. One for Bill and one for Thomas. Bill, on Ocrevus, two comments please. Your suggestions around the six-monthly cycle and the weakness in Q2 therefore having an impact on 4Q would suggest that the very strong bolus we saw in Q3 2020 should lead to a very strong recovery of Ocrevus in Q1 2021. Can you give us some indication as to whether what you'd seen in January to date supports that? I'd be very interested in your thoughts. You've said previously you're following data around COVID survival rates on patients who've had B-cell depletion, and I'd love to understand now that we've had significantly longer and a deeper pandemic, whether you remain confident that doctors are committed to B-cell therapies during the pandemic. Secondly, an easier question for Thomas.
You've set out the uncertainties but also the opportunities of diagnostics in 2021. I'd be very interested if you could get your crystal ball out and let us know what proportion of profit you think will be booked in the first half versus the second half of the year.
Yeah. Bill, before you answer, we had a similar question from Naresh from Intron Health. Only that we get everybody aligned here. Over to you, Bill.
Okay. Yeah. Thanks for the questions, Matthew and Naresh. Let's see. In terms of the six-month cycle and seeing that play out, I would just say from what we've seen so far this year, we don't see any reason to change our view on that. If we did, I would have said something different. Yeah, there's really been, since April, May, there's really been no change in the dynamic on returning patients. We have a high level of returning patients, and they come in about every six months. We haven't seen any reason to change our view on that. In terms of confidence in B-cell related therapies, let's see, our latest data point was October, where we had approximately 40% of new and switching, so that was pretty well into the pandemic.
Again, so I don't really have any reason to doubt that doctors are going to have confidence in this MOA. The safety profile has been unchanged since launch, which frankly, in itself, is a bit unusual because I think every other MS therapy that I followed, the launches, usually over time, some additional things got added to the safety profile, and Ocrevus has held up really well. Again, we're very confident in the future. There was also this question about whether there's pent-up demand, and that will we have a glut of new patients later this year? I don't know. We'll see. Naresh, I think you also asked if there was a bolus, but now we're almost four years on the market, so I don't think you could really characterize the continued strong growth as a bolus. I think it's really just patients.
The existing MS therapies have a lot to be desired in terms of preventing disease progression. When patients' MS worsens, they look for an alternative, and Ocrevus is the number one choice.
I guess the other question was for me. Having a crystal ball will definitely help in these very uncertain times. There are a number of factors that really can impact the sales positively or negatively throughout 2021. What is clear is that the growth we had in Q4 with 28%, and I mentioned that we had an opportunity to outgrow that even in Q1 specifically, but also Q2, so that we will have very strong growth in the first half year. That will translate also in strong profitability growth. Now, what's going to happen in the second half year is a question mark. One is how quickly is the rollout of vaccines. Now, is this going to take longer? The question is really on the different variants.
What kind of level of vaccine rate do you need to have to get really to herd immunity, et cetera? As this may take longer, then there is an opportunity to have a stronger second half of the year, but always against a much higher base, because if you look at this year, we did have a very high base in Q3 and Q4. For me, I think the first half year is very clear. This is where we have most of our sales in terms of COVID sales for 2021. Then second half year, we'll see how certain things develop. As the year progresses, we can inform you and update you on that.
Yeah. Thank you. There is a question from the chat, from Emily Hutchinson. She is asking for Andrew Baum. Bill, one for you, your confidence in gantenerumab going forward. I guess this is on the mind of many investors.
Yeah.
Only.
Well, believe me, it's on our minds. We have lots of reasons to believe, and we have reasons to be concerned. Until there's a definitive, unassailable pivotal result or pair of pivotal results, I think we have to handicap all these studies. That's just how it is when you're pioneering in a new area. I mean, I could give you the five reasons we're excited, and then I would still say it's, I don't know. Is it 50/50? It's something like that. I don't think we need to get precise, this is a difficult disease. We don't understand the pathophysiology of Alzheimer's disease, and we're targeting something that is in huge evidence that it's linked to Alzheimer's, but the causative role that it plays, we don't know.
The data that came out from another company in the last quarter, people got really excited about, and I want to be excited too, but again, that's a little bit of a slight tweak on the MOA. It's a novel endpoint, it's a small data set, and we haven't seen the data yet. I'm not going to take that and run off to celebrate. I think the good news is that we're coming up on a readout. It's in our sights.
Absolutely.
This day's been waited for a long time, and we've got a sub-Q formulation, which is very exciting because if you can imagine hundreds of thousands or millions of elderly people going to get their monthly IV infusion. I struggle with that one. I'm pleased that we were able to get to a sub-Q formulation, and we'll look forward to that readout.
Yeah.
Thanks, Andrew.
Yeah. Thank you Bill. Next one would be Sachin Jain. I'll allow you to talk.
Hi, it's Sachin Jain here. Thanks for taking my questions. A few, please. First, I want to just kick off with guidance. In 2021, I wonder if you could just comment on the balance of growth between pharma and diagnostics within group guide. Does pharma grow CER in 2021, and is the majority of group growth coming from diagnostics, just to get the shape of the divisional growth correct? Second question is on faricimab. The profile you've delivered, Bill, you weren't particularly excited by non-inferiority with 3Q calls. Wondering if you could just clarify the change in your thought process there and how you frame the dosing frequency advantage. Is this, in your mind, transformational to current practice or incremental and becoming a marketing battle? Then I just have one on Tecentriq adjuvant lung. You've given, in your introductory comments, timing for Polivy at midyear.
Doesn't seem like you're giving timing for adjuvant lung anymore. I wonder if you could just comment, do you still expect to be first and by a margin that's important enough commercially, as you'd commented before, or not? Thank you.
Sachin, perhaps I can just comment on the first question around the guidance, and where the growth is coming for, before I hand over to Bill. Of course, with the pandemic, there are different scenarios, and there's a certain uncertainty. I think it's safe to say that we should see a very good growth for Diagnostics in the first half of this year, and there are really two factors underlying that. One is that the pandemic, of course, still leads to big demand. Also, we have a base effect. If you look at the sales development last year, Diagnostics sales were really kicking in only in the second half of the year. There's no doubt that in the first half of this year, we will have a very strong growth in Diagnostics. We will have a base effect kicking in in Q3 and in Q4.
The uncertainty is around how big is demand and how does the pandemic develop. I think it's fair to assume that the sales growth in the second half will be lower than the sales growth in the first half. For pharma, it might just be the other way around, right? If we do see the pandemic getting under better control, patients will take up their medical appointments. Then, of course, we also have an increasing benefit of the growth of the new products on the one hand and a decreasing effect of the biosimilars on the other hand versus the previous year. I hope that gives you a bit of color how we see it evolving. At the end of the day, there's a lot of uncertainty as well.
Overall, we are confident to grow low- to mid-single digits with this kind of reverse effects in the first and the second half. Let's see where we get. Bill, over to you.
Question about faricimab. I think I've probably said most of what I can say without divulging data. We look forward to that opportunity to present the data at the upcoming retinal meeting. I think then we'll let the doctors decide what it means. I would say that we have a high degree of confidence in what we're seeing, that this does represent a meaningful step forward for patients and physicians. It's certainly made a difference in the past, the dosing interval. Now we're talking about, for example, 2x or 3x the dosing interval for many patients between faricimab and PDS. In chronic therapy, those kinds of things can make a big difference. I think there's a long history of that in some important medicines, including one that's been the largest therapy in the world.
It was all about a dosing interval. I look forward to your view when you see the data. I think you had another one about.
Adjuvant, if you will be first.
Yeah. Adjuvant lung. Yeah. Let's see. Yeah, there's no change on that, and we're confident in the timing. I think, what have we been saying? First half, Franz?
We have said that we have a chance to be first ahead of the competitors, we didn't precise when exactly it is first half or second half. We can confirm the readout in 2021, adjuvant. That we can confirm. We know that, yeah.
Yeah.
Okay. Thank you. Can I just go back to Severin on, so very kindly provided the phasing of growth, but on Roche Research. The question was more around the divisional. Does Pharma grow in 2021? Thanks.
Severin, you're on mute.
Sorry for that. No, that's certainly our ambition that we grow for pharma next year.
Thank you.
Sorry.
Yeah.
Thank you, Sachin. The next one would be Tim Anderson. Tim, I open your line now. Tim, you're on mute.
Yeah. Can you hear me?
Yeah. Now we can hear you. Yeah.
Okay. Thank you. Two questions, please. Evrysdi for SMA, you mentioned sources of switching business, both Spinraza and Zolgensma. I'm just trying to understand Zolgensma and how that would be a switch exactly. Are you saying patients forewent Zolgensma in favor of your product, or there were patients who received Zolgensma and then went on to receive your product? Then a second question on a pipeline product, tominersen for Huntington's. Are we going to see any data related to this program in 2021? What's your level of enthusiasm for this program relative to your level of enthusiasm for gantenerumab? Which excites you more?
Crystal balls today, huh?
Great. Let's see. Just to clarify, I think this is just a convention, but usually switching just means that there's naive patients who haven't had a therapy, and then there's patients who have had a therapy, and we don't differentiate between someone that had, let's say, Spinraza last week and somebody who had Spinraza last year. Likewise, if someone's been treated with gene therapy and then they go on Evrysdi, then they would be counted as a switch. Although, I certainly see your point. To be clear, we're not talking about people who forewent gene therapy. We're talking about people who received gene therapy and subsequently, at some point after that, were put on Evrysdi. The bulk of patients, as you might imagine, because many more patients have been treated with Spinraza than have been treated with Zolgensma.
The bulk of the switchers are patients coming from Spinraza. We've had a not insignificant number of patients that were treated with gene therapy coming onto Evrysdi. On tominersen, again, I sort of repeat what I said about Alzheimer's, and unfortunately, Huntington's, there's never been a therapy. It's a promising target we're pursuing. We know we have a biological effect. We have a pharmacodynamic effect on the mutant Huntington's protein. We also have an effect on the wild-type Huntington's protein. To translate that into clinical impact, it's going to require the phase III data. We thought we might have some sort of an early look. This was, I don't know, 16 months ago or so, we thought, "Oh, maybe there'll be an opportunity to see something early." We now think we're just going to have to wait for the phase III.
I don't believe there's going to be any sort of game-changing data in 2021. I think the final outcome is going to be the one that matters in 2022.
Yeah. Thank you, Bill. Thank you, Tim, for your questions. Richard Parkes would be the next one. Open your line, Richard.
Hi. Great. Hopefully, you can hear me.
Yes. Perfect. Thank you.
Just couple of questions. Firstly, on the biosimilar, looking back now, obviously the impact's been a bit more than you expected. I wonder, looking back, is there anything that you would do differently in terms of pricing or contracting strategy, if you could look back? I'm just kind of thinking about when we model biosimilar impact for Actemra and Perjeta over the next few years, is this a good example of what to expect, or are there other things that maybe you could do differently to mitigate some of that impact? That's the first question. The second question, I understand you're probably frustrated with giving a 2021 outlook and start to ask about 2022, but if you could help us to understand some of the trajectory of the headwinds as you go into 2022?
Obviously, you've got the, I think, the Esbriet patent expiry and Lucentis biosimilars, maybe a continued tale of erosion of the Legacy 3. I know some investors are concerned that diagnostics could then become a kind of headwind versus the tailwind currently. Could you just talk about the trajectory of those headwinds into 2022 and maybe your ability to maintain or improve the current momentum in terms of earnings growth? Thanks.
Perhaps I can start with the outlook beyond 2021. I think on a high level, what we have is a really huge impact of Avastin, Opdivo, and Herceptin this year as we had last year, right? Of course, there will always be products losing exclusivity. We will get out of this phase where we have such a huge effect in a very short period of time. If we combine that with the dynamic we see with newly launched medicines such as Evrysdi, if I now look at the upcoming opportunities with faricimab in particular or PDS, where we have the data. This is not speculation anymore. We know what the clinical benefit is, and we are very confident about the opportunity here.
If I put all of that together, the dynamic of the newly launched medicines and how the portfolio is ramping up, then I think it's safe to say that we will see an acceleration of growth into 2022. The question then is really of whether some of the really big opportunities and risky opportunities materialize, such as Huntington or gantenerumab. That, of course, would be transformative. Even if those do not work out, I think we have the worst behind us in 2021. The worst is not so bad after all. Who would have thought a couple of years ago that we can grow through this biosimilars erosion phase? Yes, we are pretty confident beyond 2021. Bill?
Thank you.
Yeah, I would just add to that on a couple of things you mentioned, for example, Lucentis biosimilars are coming, so we will also have the port delivery system with ranibizumab launching, and we'll have the faricimab launching shortly behind. Lucentis is really quite a small product now. Esbriet is obviously much smaller than the things that we've lost. The Actemra biosimilars, our understanding is those have been delayed, and so that's out some time. I think, when you look at the things that are coming, whether it's tiragolumab or potential for gantenerumab, just go down the list, I think we feel quite good about our long-term growth prospects. Let's see, you asked about alternate actions or the things we would have done differently. Yeah, it's not obvious.
The main thing you could do is you could try to compete on price, frankly, what we see in the biosimilar realm is very, very deep discounts. For example, in Europe now, there are discounts, 85%- 87% discounts. You can imagine, it's pretty hard to win in that kind of a race. I think, we feel good about the actions we've taken, and again, our real focus is on innovation and out-innovating ourselves.
Yeah.
Thank you.
Thank you, Bill. Next one would be Wimal from Bernstein. Wimal, I open your line now.
Thanks, Karl. Wimal Kapadia from Bernstein. Bill, I just want to ask a little bit of the upcoming catalysts, just to get a sense of your level of confidence, and in particular, Polivy in first-line DLBCL and Tecentriq in adjuvant. Particularly tied to the former, how quickly do you actually think the standard of care can change if we see success? Docs have been using R-CHOP for quite a while, like you highlighted. How realistic, if we see success, could you see a relatively rapid transition? Just your level of enthusiasm for both of those two readouts. Can I just ask a little bit about pharma in China, please? You saw a strong uptake of Perjeta and Alecensa now partly offset by impact from the NRDL.
Can I just ask a little bit more about the dynamics of the biosimilar exposed assets and Tecentriq in the region? How are you really thinking about 2021 for pharma in China? Thank you.
Yeah. Thanks for the questions. First, in terms of Polivy, we had some really exciting data in the preclinical models that led us to take this molecule forward in the first place. The efficacy is really outstanding. We really believe this is the best opportunity we've seen yet to advance on what Rituxan does in first-line DLBCL. Beyond that, again, that's why we do the phase III study. There's not an opportunity to see it in this type of setting, to see that kind of impact. I think this is kind of like the Cleopatra data for Perjeta or something, where you're going for the adjuvant win. This is like that. We're going for cure. It's in first line, and we'll know when we see it.
I believe that the standard of care would change relatively rapidly because, you may recall, when Cleopatra was presented, people looked at that and said, "Well, how big of an increase is it?" "It's an extra medicine," we were confident always that the standard of care would change, it's really simple. You're going for a cure, these are younger patients, everybody can see the difference between curing 75%- 80% of people. That's five people who you're saving. I think that we won't have trouble convincing people if we have the data. Tecentriq in adjuvant, I think we're quite enthusiastic about that. It's true that in lung cancer, there's not as much adjuvant treatment as you have in, say, breast cancer, because unfortunately, typically lung cancers, you don't catch them until they're metastatic.
The market size is not as large as, say, adjuvant breast, but it's definitely a blockbuster potential and a very meaningful market size. We've run a really good study. I think we have a great opportunity to show a positive benefit if cancer immunotherapy is going to matter. There's a lot of reasons to believe it should, because you are looking for sort of micro-tumors that are not detectable and an ability to control those or wipe those out, and there's a lot of reasons to believe that a cancer immunotherapy would aid in that. Let's see. You asked about China. There's a tremendous unmet need in China. There's a huge population to benefit, and we've had this dynamic where we had to reduce prices in order to get into national distribution and really have all patients eligible.
That has an immediate effect on your sales, because if you cut price by 30%, your sales go down the next day by 30%. So you've got to grow sales by 50% to make up for that, and it takes some time to do that. The pandemic hurt us this year because we cut prices on some important products at the end of 2019 in order to get on the NRDL, and then there was a lockdown in Q1 across China. As you know, and you've no doubt heard, that has left a lasting impact on the China health system in terms of their capacity. That said, we saw a nice recovery by the end of the year. We had solid growth in China for the year, I think it was 8%. Solid.
That was despite the fact that there are local competitors to AH&R, but we still had very strong volume gains on those products despite the local competitors. I think, China's going to be a different kind of market than, say, Europe or the U.S. I think there's still a lot of room to penetrate on volume, and we will have some price challenges. I think, Did you ask about Tecentriq?
Tecentriq, yeah.
I think this is one where there's a number of local competitors, checkpoint inhibitors, the market has been really substantially deteriorated. I think we have to take that into account in our outlook.
Thank you. Thomas, can you also quickly comment on China for your business, for diagnostics?
Yeah.
Exactly.
Yeah.
Thank you so much.
Yeah. Thank you very much for your questions. Thomas, maybe you could kindly comment because we got one online question here which-
Yeah
which goes in that direction, yeah.
Yeah, happy to do that. We did have a significant impact on our core business, particularly in March, April, May in China. The sales in China did recover, and if I just look at in-market growth, so not what we sell to distributors, but what our distributors sell from our product into the market.
it was already in the high- single-digit range. You don't see that because we also lowered inventories from 80- 45 days. Yeah, we'll see strong growth in China in this year. One, because we have winds in our back and some tailwind because of the lowered inventories, also because we see in-market growth. I do believe we will have good growth, and yeah. We're seeing that already.
Yeah. Thank you. Thank you, Thomas. Next question would be from Emmanuel Papadakis from Deutsche Bank. Emmanuel, I open your line now.
Karl, thank you.
Yeah, we can hear you. Yeah.
Thank you for squeezing me in and taking the questions. Maybe I'll take a follow-up to Thomas on Diagnostics side, just to try and get a bit more of a sense to which the extent to which the growth may be durable there. Obviously what we've seen so far is very much driven by molecular biology and both the point of care and antigen testing. What extent is that going to translate into a more durable expansion of the instrument footprint? Far, it looks like it's pretty much concentrated in those two segments, which really correlate with the prevalence of immediate COVID-19 diagnostic testing. Is there going to be a consequent durable uplift in your view for, in particular, for example, the clinical centralized lab business? That's question number one. Maybe I'll take one, perhaps for Bill on the R&D side of things.
It's nice to see some of the Spark assets coming through the pipeline. Perhaps you could talk a little bit about timelines and intent on the enzyme replacement gene therapy side with the phase II Pompe initiation. Thank you very much.
Yeah. Happy to take the first question.
Yes, please. Yeah. Mm-hmm.
Yeah. We definitely see very strong testing, particularly first half year, likely some testing also the second half year and for the years to come, simply because this virus is endemic. We will have to deal with it. It mutates frequently, so we will have to monitor this. With regards to the level of testing, that's very hard to predict, but definitely a lot more than what we would see today with influenza testing. It may even have an impact on other tests like influenza testing, because simply they want to know what kind of respiratory disease is it, and just scan that. With regards to beyond that, in the outlook, first of all, we made extreme progress in expanding our menu on the 6800, 8800. There's still a number of tests that we can bring onto that platform. That's fantastic.
Also we are installing in one year what we have installed the prior five years. We are more than doubling our installed base out there of the systems. There is a huge opportunity, and I just want to pick three. One is around cervical cancer. There are more than 300,000 women that die every year of cervical cancer. Many countries still do Pap testing, which has a sensitivity of 50%-60%. Zur Hausen almost won, 20-30 years ago, a Nobel Prize showing that HPV causes cervical cancer, and still many countries don't do it. While they can save 300,000 women's lives every year. Then, I don't understand, they do so much around COVID, which is great, but they forget that a lot of people are dying of other diseases, and they should do this. Next is hepatitis C.
80 million people in the world have hepatitis C. Hepatitis C leads to liver cirrhosis, to liver cancer.
Yeah.
Many people are dying every year. There is a cure for Hepatitis C. If we screen all the people, how much suffering can we take out of the world? Just looking at these 80 million people that, at the end will suffer, their families will suffer, and the healthcare system will suffer because they have a lot of money that they need to spend to help these patients later in their life. Tuberculosis, 1/5 of the population has infection of the bacteria of tuberculosis worldwide. Look at how much we can do. We need to start to recognize what healthcare systems can do by intervening much earlier. Yeah. I have to say, there's such an opportunity and governments need to get going on this.
I'm sorry to get a bit emotional on this.
No, I imagine-
I have been fighting for 10 years with governments to include HPV screening. All the clinical data is out there, and they need to get going and not just let it go like they have in the past.
No, absolutely. There are, let's say, durable cases or use cases clearly beyond COVID, and I think Thomas made it very clear and illustrative that there is a lot to do in the healthcare systems. Thank you, Thomas. Bill, could you kindly-
I think from an economic point of view.
Okay
Just to add, yeah. If you go into screening programs, part of the problem is the upfront cost, right? You only have the benefits in the longer term. That typically is a political hurdle because politicians, at least in our Western democracies, are very linked to election cycles. Therefore, they hesitate to invest when they are not yet sure whether they will be reelected, and the benefit might only come in the next period. What should help here, and what should give a bit of a momentum is, that now the investments have been made, right? These molecular diagnostic platforms are out there. This is considerable investments, which are now done because of COVID-19, because the political pressure is also so high.
That should help to reduce the hurdle to now also use these platforms for other parameters like HPV or HCV or tuberculosis. Actually, these are all assets which you can run on exactly the same platforms. Then I think there's also an economic argument. Rather than having those platforms sitting idle in the cellar, use it. It's good for people and actually it's a lot of cost savings in the long term. The data are so clear. I think there is an opportunity to have a structural shift in the diagnostics business.
Thank you.
There was a question about the Pompe disease and gene therapy, and we're really pleased at the progress that Spark is making. Yeah. I think it's so far quite a good match because we've been able to provide significant funding for their pipeline and they're able to leverage some of the global resources. For example, in Pompe disease, they've just had their first patient in the phase II. Now, in Pompe, because there's available therapies, the gene therapy will probably have a higher standard than you'd have in some of the monogenic disorders. We do expect a full phase III there. It is going to be some time yet, I think beyond 2023 for launch. I think the phase III would be out sometime after the phase II. It's going to be some time, but yeah.
A lot of good progress in Philadelphia and we're excited about our new colleagues.
Thank you.
Thank you.
Thank you, Emmanuel, for your questions. I suggest that we give 10 more minutes because we were running over a bit with the presentation time, if that is okay for you, Severin, Alan, Thomas, Bill. Okay. It needs a bit discipline from both sides, from the investor side and from our side now with the Qs and the answers. Next one would be Luisa Hector. Luisa, please. I open your line.
Thank you, Karl. Can you hear me?
Yes. Perfect. Thank you.
All right. Thanks, everyone. One was a point of clarity on the guidance for the biosimilar impact this year of the CHF $4.6 billion. Is that still U.S., Europe, and Japan? If so, what's happening with the international markets for Herceptin and Rituxan? The second question, just to understand a bit more on Evrysdi. It's off to a good start. Just wanted to know, can patients start this drug literally from home, or do they have to come in for a first dose kind of clinic appointment? Have you lowered your expectation for sales this year given the COVID environment, or is this more of a unique launch due to the high levels of awareness of the patients and the doctors? Thank you.
Yeah. Thanks for the questions. Easy. Biosimilar guidance, the $4.6 billion is the whole world. Before we were talking about a specific set of territories, because those were the ones where the biosimilar impact was happening. Now, essentially, there are biosimilars in many places. We just said we'll handle them all together henceforth. That's the $4.6 approximation. Let's see where it ends up. Evrysdi absolutely can be started from home. It's a simple oral formulation, liquid formulation. That's not an issue. No, we haven't lowered the sales forecast based on COVID. We really don't see any need to, with more than 1,000 patients starting in the first 4.5 Months and 350 physicians prescribing. These are motivated people, and for many, well, you can imagine for the parents of young children, they're very motivated.
For people who are older, who haven't had good options, they're very motivated as well. Yeah. I think it should be a good launch in the U.S., or a continued good launch in the U.S. and strong launches outside the U.S. as well.
Thank you.
Thank you, Bill. Thank you, Luisa, for your questions. Mark Purcell. Mark, I open your line now. Mark, you're on mute. Now you're on mute. Go ahead.
Thanks very much. Just following on from that last question on biosimilars, what would you recommend us to do when it comes to modeling the decline in revenues in international markets? I guess, we don't have much visibility there. We obviously have now high visibility what happens in Japan and the U.S. and Europe. Whether you're going to see sort of flattish sales or slightly declining sales as prices come down, but volumes obviously go up in those markets. Secondly, bispecifics. Bill, could you talk to your confidence in terms of this modality where you have 15 assets being able to be delivered subcutaneously in an outpatient setting? I'll be interested in the comments around the CD20/CD3s, but also where you are with HER2 CD3 in terms of gaining share back through innovation in the breast cancer space.
A very quick one, which is a clarification one for Thomas. Thomas, thanks for your presentation. It was really helpful. In terms of thinking about diagnostics through the course of 2021, without giving a growth rate, for your own business specifically, are you assuming much in the way of pricing pressure or capacity utilization pressure in the second half? Are you playing a part in what could become new tests, depending on what the correlates of protection are with these COVID vaccines? We may be shifting from nucleocapsid antibodies to total antibodies, to T cells, to all sorts of other things. I would have thought with your innovation focus, you may play a leading role there, which again, could give you a little bit more longevity when it comes to the COVID revenues. Thank you.
Okay. Maybe I could start with the question about the biosimilars impact in international. Yeah, Mark, it's a little hard to call because there's definitely some growth potential. Some of this will come down to what are pricing impacts in 2021 and future years. I think certainly the great bulk of the biosimilar erosion will occur in the U.S., Europe and Japan. Again, for example, China's in international, and we have every reason to believe that we would have some growth in China this year for AHNR. I don't know that I can help you much beyond that. We're going to have to wait and see.
On the bispecifics and the encouragement or the potential for sub-Q dosing, yeah, I think we're pretty impressed with the data that we showed at ASH in terms of the ability to administer Mosun, to basically be able to give a much higher dose with sub-Q but have it be well-tolerated. I think this is something that could be quite useful for a number of these Well, in this case, I think that's really more specific to the hematology indications where you have CRS as an issue. It's something that we think could be an important phenomenon in hematology. You asked about the HER2 CD3 bispecific. It's in phase I. We've had some delay on that one based on COVID. We look forward to further progress on it. But right now it's moving a little bit slowly.
Let me answer your second question with regards to pricing pressure, volume, et cetera. As I mentioned in the presentation, most of our outlook is really having COVID sales in Q1 and Q2, and then a reduction in Q3 and Q4. This can either be pricing pressure or also difference in volumes, and it will all be dependent on the speed of the vaccine rollout, the variants. With more of the variants of South Africa and Brazil, most likely you have to get a much higher rate of vaccination in order to get to herd immunity, because with the higher infective virus, you need to have a higher level of vaccination. There is a lot of uncertainties that are going to play in. We didn't really detail out what part of it's price pressure, what part of it is volume.
We said, okay, first half year is pretty clear. The second half year is a bit more unclear, so we take more of it in the first half year, and then we can give updates as the year proceeds. I also think that we are pretty well-protected compared to some of the other players, both on price and on volume. Price, because we do have a lower price than most other players, and significantly, partly. If there's a price erosion, others will be hit first before we will be hit. We did price in line with our other portfolio, right? It's not like we priced higher. On the volume, I also think we are somewhat protected because our machines and instruments are so highly automated that all these more manual procedures will go away because people are just tired.
They don't want to keep pipetting and work on these more manual instruments. I think there's a certain level of protection. With regards to more innovation in this space. We do have now an opportunity also on the MagNA Pure and LightCycler to be able to differentiate in this new variants already. We are working on something on our 6800, 8800 that you can then take basically the samples that turned out positive, and you can kind of look, is this a new variant or is this the previous virus? That through PCR rather than sequencing, because sequencing just takes days versus hours of PCR, and you don't really want to wait that long. Of course, this question around T-cell immunity, and there are possibilities to test that, and we are working on that as well.
We do believe, however, that most of the testing will still remain on antibody testing, because that will show also correlation to this T-cell immunity. This T-cell testing is quite more cumbersome. We do look into that and are working on solutions there as well. I think, if they're looking at this testing, they will do more of the antibody than the T-cell testing, simply because antibody tests, antibodies presence will correlate to T-cell. The T-cell is just a lot of manual effort.
Thank you, Thomas. I can see that your crystal balls or let's say, forecasts here, August are really in high demand. We have a bit of a timing issue. We have Richard Vosser and Peter Welford as the last two identified, let's say, people in the queue. I would say, let's give it a try. Two quick questions from Richard and from Peter, and we try to also keep ourselves in a good shape with the answers. Richard, you first, please. Richard, I open your line.
Thanks, Karl. One question on diagnostics. You mentioned the ramp-up in capacity, what is the demand level like? We've seen PCR selling out this year, is there the demand to sell out of that ramped-up capacity in the first half? Maybe an idea on the sustainability of margins. We could expect margins to go up, would they be sustainable at sort of a 20% level going forward? How should we think of that? Very quickly on Tecentriq, just adjuvant growth. I'll stop there, Karl. I heard the sigh.
Thank you. I already can hear that's good. Thanks a lot. No, please go ahead, yeah.
Regarding the ramp-up capacity, the numbers that I mentioned are always at end of quarter. Don't take this as during the entire quarter. This is doubling by end of quarter one and end of half year, another doubling. Now, since there is so much demand out there and not enough supply, it's really hard to say what is the real demand. Based on all of our models, it's likely that the demand will be higher than what we can supply still, despite investing more than CHF 600 million. We'll see how second half year goes, right? Regarding sustainability of margin, yeah, we've had a significant margin increase, also driven through COVID tests in 2020.
As you have seen on the cost lines, we're working very hard on our efficiency in the organization, and we'll continue to do that to make sure that we protect our margins.
Thank you, Thomas. The last one would be Peter. Thank you. I hope we could address your questions, Richard. Thanks for asking. Last one would be Peter Welford. I'll open your line now. Thank you, Peter. You have the privilege of the last question.
Hi. Thanks so much for squeezing me in. I just have one question, keeping it short, just to Thomas again, though, just continuing on the theme of the outlook, actually. Just want to check there. When you say you assume little testing in the second half relative to the first, is that both antigen and PCR tests? When we consider the PCR then, am I right in thinking you said you're at 20 million tests a month at the moment, 40 million by the end of the first quarter, presumably then 80 million by mid-year, if I understand it right? When we think about the second half, you're saying your guidance assumes down from 20 million a month, essentially. Actually the upside, if you like, could be as much as 80 million a month tests. Is that the right way of thinking about it?
Thank you.
Yeah. By half year, I would say, to be more accurate, it will be around CHF 70 million in total. Right. It's a mix of our highest super platforms, which is CHF 60 million, and the other CHF 10 million come from the MagNA Pure and LightCycler platforms. It is correct that we assume lower testing at the moment in our outlook for the second half of the year. The other view that I would just like to bring in, we believe that PCR testing will be sustainably longer, will be longer sustainable than, for example, antigen testing. I think one of the reasons for antigen testing also is because there was not enough of PCR testing. I believe that the PCR volumes will remain longer than antigen.
Yeah. Thank you. I think we are coming to an end. I apologize with those who are still in the line. Keirnan, sorry, see that you just came in in the last moment. I see that Ben had some questions on ESG. We had some further questions from Michael Leuchten and some on the digits and so on. We take note of it. We'll get back to you. I apologize again for those whom we couldn't address here. I want to do thank you from my side for your interest in Roche. Wishing you health and success, health for you and your families of course, and wishing you a nice day. Thanks a lot for your interest.
Thanks.
Thanks, everyone. Bye-bye.
Thank you. Thanks for joining.
Thank you.
Thank you.