Christian Hogg - CEO and Executive Director.

Susie Jana - Edison Investment Research Limited Christian Glennie - Stifel, Nicolaus & Company.

Okay. Okay, I think we've got everybody. So welcome, everybody, to the 2017 Full Year Results Presentation for Chi-Med, for Hutchison China MediTech. What I'm going to try and do is take you through the presentation quickly maybe in the next 30 or 40 minutes, and then leave ample time at the end of the presentation for questions, for Q&A.

You'll see that from this presentation, it's really an update. I mean, obviously, the 2017 results are broadly explained. But each of the clinical trial programs, it's an update. So I'll take you through the changes since the interim results six months ago. So overall, the momentum continuing to build strongly for Chi-Med.

We've - we had a fantastic 2017 pretty much by all metrics in the context of the performance of our business as well as the progress of our clinical drug candidates. The first NDA submission went in, in June of last year. We've been navigating the approval process over the last six months.

We'd hope it would go faster, but it's a laborious and very detailed process that I'll give you a bit more of an update on later. But we still expect probably around the middle of this year should be done, but subject to SFDA activity and actions on their part.

In the context of what's going on at the moment, we're now into or just entering the final stage of inspections, and that's around our good manufacturing practice formulation facilities as well as the GMP facilities, where we produced our APIs.

So - and all the clinical reviews have been done, all the clinical site visits and inspections have been done, all of that work has been done.

It's now the final part, which is the manufacturing part, which we spent the last three years building our manufacturing facility with Eli Lilly using their global CMC organization to help guide us on that, so we don't expect to have any issues with that. So that's fruquintinib, target to launch this year, and that'll be our first launch.

On savolitinib, we've put a lot of information into this latest announcement today around non-small cell lung cancer, the Phase II results that were presented at World Conference on Lung Cancer in October of last year, and AstraZeneca's decision to proceed now into the last stage of development on savolitinib and Tagrisso combination.

So we can go through that in a bit more detail later. But overall, we have probably the biggest development program that we as a company have ever been behind that is about to take place on savolitinib and Tagrisso in terms of scale, global reach as well as the financial cost of it. We have six Phase IIIs underway and completing.

We just announced recently the FALUCA study in third-line non-small cell lung cancer on fruquintinib, completed enrollment of 527 patients in China. We'll report top lines sometime, probably in the fourth quarter, I would imagine mid fourth quarter of this year. So before the end of this year, I would expect top line. I mean, it might be later.

But if it is, it's because OS is maturing better than we expect. But I would imagine it will be somewhere in the fourth quarter. And we have 22 Phase Ib or Phase II proof-of-concept studies underway on eight drug candidates. I mean, we are, today, enrolling patients in 36 target patient populations across the world on our eight drug candidates.

So it's a very large range of clinical activities that we're involved in. Weiguo is here, our discovery engine continues to produce. We haven't made any announcements yet, but internally, we've agreed candidate nomination on a - on novel drug candidate that is now moving into regulatory tox testing.

We'd hope that, that would make it into the clinic next year. So we're producing 1 to 2 novel drug candidates a year, with an organization of around 200 discovery people in China. Fully integrated. All of those programs that we have in the clinic today have been generated in house, which is quite different for many Chinese companies.

Most Chinese biotechs these days are licensing in from outside of China to bring innovations into China. We have created over everything ourselves. And so that's good in a couple of ways because you design these drug candidates to do what you want them to do and to the standards that you are requiring.

The new wave of innovation is primarily in the immunotherapy area, second-generation I/O targets. Won't talk much about them at this stage, but they will play out over the next year or two.

And then the Commercial organization, I'll take you through the results, but our Commercial organization this year - or in 2017 and continuing into '18, we've had a very strong start in 2018 on our Commercial organization. We've gone through a major change, building two enormous GMP factories in China because we were capacity-constrained.

And now we're starting to see the benefit of unrestrained capacity. And so the 2017 results, we shall take you through in a moment, were spectacular. And as far as the - not just the size and profitability of it, but also the capability of it is very strong.

We'll give some examples on Seroquel and Concor about taking on new products and commercializing them, and we've just made great progress on these third-party products. So overall results, in line with guidance. The guidance was $13 million to $28 million.

It was quite a wide range last year because we had the land compensation in there, which was sort of zero to $13 million of land compensation. That didn't come.

But it - even despite the land compensation in Guangzhou not coming, we still met our guidance because of the strong performance of our China business, which - I don't know whether you can see it here, maybe we'll wait onto the next slide, but we reported a net profit after tax, excluding onetime items of, $37.5 million, which was up 25% versus year before on our core China operation.

The investment in the Innovation Platform was about $52 million net loss. And that, in total, was about $88 million in non-GAAP investment expenses on our pipeline. So overall, in line with what we'd predicted. China being particularly strong, and Innovation side increasing its spend. This chart shows it a bit better.

On the Innovation side, income and revenues about the same as the year before, around $36 million, a net loss of about $52 million, leading to about $88 million in innovation investment. And then the Commercial side, $37.5 million net income as compared to $29.9 million the year before, so a 25% increase. So that was very positive.

Looking forward and looking at the cash situation now, we are sitting on, or at least at the end of the year, sitting in around $480 million in cash and unutilized banking facilities. The unutilized banking facilities are about $121 million of that. The cash sitting in either short-term investments or in the bank is $358 million.

We have about $30 million in borrowing. We had a successful follow-on offer in October of about - around a little short of $300 million met. And our JVs continued to be well cashed up, $67 million in cash and no debt in our JVs.

So from a cash standpoint, we have all the resources we need to move our pipeline through what I believe is probably going to be multiple approvals over the next 2 or 3 years. On the guidance, we'll continue to progress as we have been. I think on the Innovation side, we'll see our revenues increasing as fruquintinib is approved and launched.

There are some big milestones, and then we'll start to get our royalties coming in, so you'll see an increase in revenues on the R&D side. But you'll also see an increase in the spend.

I mean, we're targeting somewhere in the range of $110 million to $120 million in investment on our pipeline in 2018 as compared to about $88 million, as I said last year. So it's a decent step-up. On the Commercial side, there are some machinations going on, on sales. We have the Two-Invoice policy in China that has come in over the last six months.

What that - it's a policy from the CFDA that basically cuts out the middleman in China. So we were performing the middleman role for Seroquel in China, buying product from AstraZeneca and then selling it on and booking sales in doing that.

Under the new Two-Invoice policy, all we're able to do is charge AstraZeneca for our services, so we charge them fee-for-service. So the organization is the same. The team is as big. The operations and what we do is exactly the same. It's just the way that we account for it is different.

The net profit, at the end of the day, from all of those activities is exactly the same, just we don't book the sales, but the net profit doesn't change. And so you can see that in Hutchison Sinopharm, in the consolidated sales of that joint venture.

They dropped from, in this case, in 2017, it's not on this chart, but they were about $166 million in 2017. In 2018, they'll be about $80 million - $75 million, $80 million. So they'll be cut in half, but the profit of Hutchison Sinopharm will actually increase in 2018. So that's one thing.

And then I think if you look at net income for the total Commercial Platform, that will just continue to rise regardless of any changes in how we account for the provision of these services that we provide. So you can see from $37.5 million up to $41 million to $43 million is the guidance for our Commercial Platform next year.

That's about 10% to 15% growth, which I think we should be able to do pretty, pretty easily. And then you've got a big gap on the onetime gains. This zero to $20 million is all around the land in Guangzhou.

We're working very hard with the Guangzhou government to actually transact this compensation, to get the compensation for this piece of land we have in Guangzhou that we're about to give back. And it may well come this year or it may come next year, so we have to put in sort of zero to $20 million on that.

So that's the picture for the guidance is that - it's a pretty broad range of guidance there, but that's because of some pretty big factors. But underlying, what you've got is continuation of what we've been doing.

Number one, increase in the activities on our Innovation Platform; broadening of investment on our clinical pipeline, underpinned by increased income from our partners, AstraZeneca and Lily; and increased growth in our Commercial Platforms. So that's basically the story.

News flow or potential milestones in 2018, there's a lot going on in 2018 on this portfolio. Savolitinib, you can see the initiation of the global, randomized, chemo-doublet controlled study of savolitinib and Tagrisso in combination in second-line non-small cell lung cancer. That's going to start up this year. It's currently under planning.

It's powered for ORR and PFS. It starts out as a Phase II. It doesn't start out as a Phase III because we haven't yet spoken to the regulatory authorities and got the design of this study all agreed, but that's going to happen in due course. We continue to work on the dose.

The Phase Ib/II study that was presented at World Conference on Lung Cancer was at 600-milligram once-daily dose, and that's the core dose. We have agreed with AstraZeneca now that we plan to use a weight-based algorithm for dosing. So heavier people will get 600 milligrams, lighter patients will probably get 400 milligrams.

But there's a lot of scientific rationale for that it. It's actually the dosing algorithm we use in the papillary renal cell carcinoma study, so that's good.

I'll show you a little bit later some data around the tolerability of the savolitinib-Tagrisso combination, how that compares to other treatments in non-small cell lung cancer, and that how that compares to other targeted therapies. Basically, the savolitinib-Tagrisso combination is very well-tolerated and safe combination.

And - but looking forward, this work on dosing that we're doing at the moment and enrolling patients in TATTON D at the moment is going to help us essentially define the dosing regimen that is going to provide for patients to stay on this combination for the longest period of time.

Because some of these patients in our early studies have been on this regimen for a very long period of time, and you want to keep these patients on the drug. So that starts out, plus multiple supporting studies.

So the main randomized - global randomized, chemo-doublet study starts out this year, but there are multiple tolerability supporting studies in TATTON D as well around the dose that are all going on at the moment, and will - some of them will start up during the year. Third-line non-small cell lung cancer, sort of post-Tagrisso.

As all of the second-line program, number one, plays out; the third-line programs will also play out. Astra is pretty excited about that. Tagrisso is $1 billion drug now in its second year of sales.

So savolitinib coming in as a perfect bed fellow with Tagrisso, it really is strategically very attractive to our partner, AstraZeneca, to broaden their franchise in this area. Number three is the pivotal study of savolitinib and Iressa in China. This is one that we are most interested in doing. Astra, I would say, are a bit less interested in this.

And they're less interested in it because they think it might cannibalize the Tagrisso-savolitinib combination, so I can understand that. But from our standpoint, this isn't about Tagrisso cannibalization, this is about maximization of savolitinib commercial opportunities.

So we're pushing very hard to go ahead with that Phase III of savolitinib and Iressa in China, where there's an enormous unmet medical need in EGFR mutation-positive non-small cell lung cancer. So we're working that. You'll see in all the documents that we've published today, it's a discussion. It's not a decision, it's a discussion that's underway.

And I'm sure the relationship we have with AstraZeneca is a very positive relationship. I'm sure we will solve that discussion and conclude that we should go ahead in one way or another on that program. And then the molecular epidemiology study in papillary renal cell carcinoma, that should complete and be reported around the end of the year.

And our hope is, on that, that you're able to really show that c-MET is a major negative prognostic in papillary renal cell carcinoma. And as a result, take our Phase II data, the molecular epidemiology data, and go to the U.S. FDA and talk to them about Breakthrough Therapy designation on savolitinib in that patient population.

Fruquintinib, as I've said, the NDA approval and the launch in China that should come all this year, and then the release of FALUCA data in third-line non-small cell lung cancer. And then epitinib, finally, the start of Phase III this year.

We published at World Conference on Lung Cancer in 2016 the epitinib Phase Ib/II data that was really encouraging. We've been working during 2017 also on dosing. There is a cosmetic sort of AE that emerges on skin pigmentation around use of epitinib at 160 milligrams a day.

It's just - and actually, a lot of the patients of these non-small cell lung cancer patients, EGFR-driven non-small cell lung cancer patients are women. And they've got brain metastases, but they still care about the pigmentation of their skin. So we've been working on 120 milligrams. And where we are is we've got very strong target coverage.

Got great efficacy at 120, we don't have the pigmentation issue at we have at 160. So more likely, we'll go into that Phase III at 120 milligrams, and we've taken a good sort of 6 to 9 months to figure that out. But so we're quite keen to get going on epitinib, and we will do this year000. Sulfatinib, the Phase II in the U.S. is starting out.

523, the Syk inhibitor, we're targeting ASH at the end of this year to be able to come with an update on - in hematological cancer patients on the Syk inhibitor. We're encouraged with the early efficacy that we've been seeing on that. But by the time we get to ASH, hopefully we'll have significantly more data.

And then 689, the PI3K delta, that's moving along nicely in China now, and we're also encouraged by what we're seeing. So the team is now up to 360 scientists and staff in Shanghai and Suzhou, close by Shanghai. We've put in well over - now over $500 million of investment into this platform, and the results are as you see.

Chemistry is our edge, I won't go through this in a lot of detail. You all know this. And ultimately, it's not just chemistry. Our organization is very deep in all functional areas of drug discovery and development.

But I would say in particular, our technical expertise in the field of chemistry has allowed us to focus on really taking our time to design highly selective small molecule targeted therapies. That, as I will show you in the coming pages, are uniquely tolerable and - than first-generation multi-kinase inhibitors. This is a new chart.

It's very detailed. But it shows, if you look at it maybe after the presentation, it shows that for first-generation multi-kinase inhibitors, what you're seeing here - so taking Sutent as an example, it is total discontinuations of Sutent. And this is in clear cell renal cell carcinoma.

Total discontinuations using it as a monotherapy is around 35%, 40%. So this is 35%, 40% of patients just can't tolerate it for whatever reason, whether that's adverse events emerging or whatever other reasons. So that's a fairly, fairly common number for a targeted therapy, first-generation small molecule targeted therapy.

You can see some of are better, so axitinib is 17%, lenvatinib is 31%. Pazopanib, a very successful first-line in clear cell renal carcinoma drug, 36%. Sort of somewhere in the sort of high teens all the way into the high 30s is common for a monotherapy, for single agents.

So you look at the bottom five down here, savolitinib, fruquintinib, sulfatinib, epitinib. And you're looking fruquintinib, 11%; savolitinib, 14%. What you're in is, you're into sort of low teens for our highly selective targeted therapy.

So you can't - you've got to be careful comparing one indication versus another because you can see actually for third-line colorectal cancer, it's 19% for fruquintinib discontinuations. But in non-small cell lung cancer, third line is 11%.

So you've got to be careful comparing indication-to-indications because third-line non-small - sorry, third-line colorectal cancer patients are very sick patients, have a hard time tolerating anything. And in lung cancer, it's a bit easier, but that's the key here, is highly selective small molecule-targeted therapies will be more tolerable.

That will allow for patients to stay on the drug longer, for you to cover the target longer and for you to deliver better efficacy. First-generation targeted therapies hitting multiple targets with a lot of off-target toxicity are more difficult to tolerate, even on their own, let alone getting into combinations.

And we'll cover combinations in a moment. The pipeline, I won't go through it in a lot more detail, but we have laid out in our presentation for the 36 target patient populations and in our announcement enormous detail on each of those target patient populations, where we are today and what's going on.

But the - overall, we dosed about 700 - over 700 patients in 2017. I expect that will increase this year. So we're enrolling an awful lot of patients into our studies annually as our programs increasingly become late stage and increasingly advance and broaden. This chart shows the sort of 8 to 10 pivotal shots on goal.

The highlighted circle at the top is the randomized, chemo-doublet, controlled study expected to initiate in second-line non-small cell lung cancer this year. You can see colorectal cancer, that's obviously checked, but we're waiting for the NDA approval.

And then the sort second circle is fruquintinib, the Phase III top line results in Q4 of this year. So ultimately, for us, it's about broadening this late-stage portfolio and moving it, moving it rapidly on. Savolitinib, I won't go through it in a lot of details, but you can see, most of you have seen this chart before.

MET is just relevant in many solid tumor types. Interestingly, we've added prostate cancer to this table because we've started a Phase II study in prostate, MET-driven prostate cancer in Canada.

It's an investigator-led study, but it's a large investigator-led study, 500 patients, of which savolitinib is 1 of 6 arms in MET-driven patients in this population. So MET is inherent in so many places.

And ultimately, for us, it's about get it approved in lung, get it approved in PRCC, but then as we're doing that, explore all these other solid tumor indications and then go after them later as we go forward. You can see these charts are the same as we've had before, but they are updated.

You can see on Tagrisso here, last year, in just in the second-line setting, Tagrisso did just short of $1 billion in its second year of sales. I think Astra is increasingly putting out estimates of $3 billion, $4 billion in sales for Tagrisso. But I think that would mean that it would have to come into the first-line setting.

But patients failing on Tagrisso, there's a lot of MET-driven disease once patients fell on Tagrisso. And you can see in the third line, post-Tagrisso failure, MET is prevalent, MET gene amplifications prevalent in about 30% of patients. This is the big area that we're now moving forward on.

In second line, you can see 55%, 60-odd percent objective response rate. You can see, we've put out very limited duration of response data. Here, you can see on the MET positive T790M patients - T790M-positive patients, 9.7-month median duration of response. That's only four patients, so you've got to be very careful with data like that.

But obviously, since we presented this in October, that PFS and duration of response data has continued to mature and is, I think, the basis of why AstraZeneca is making the move that they're making now on savolitinib and Tagrisso because we're encouraged by what we see.

And that data will be presented, I'm sure, at some scientific events later in the year. So moving ahead in the second-line setting, where we have really a very, very strong signal. This is the Iressa combination data. You can see 50-odd percent response rate in T790M negative MET-positive patients.

As you would expect, there's very little signal in MET-positive T790M positive because Iressa doesn't address T790M positive. But in the negative patients, that is a very strong signal.

And we feel that in China, with Iressa now being off-patent and literally going from - two years ago, Iressa in China was $2,700 a cycle a month, now generic gefitinib is maybe $400 a cycle. So Iressa's cheap now in China, relatively speaking.

And so for these patients that need a combination of an EGFR inhibitor plus savolitinib, this is the best combination economically for those patients in China. If you have - if you're T790M positive, you need Tagrisso, and so that's a global opportunity as well.

Just more data on the Tagrisso combination and 30% being MET-driven, strong signal in these patients to 33% objective response rate, confirmed objective response rate.

The thing about these later-stage patient is they're very weak, and many other genetic drivers of cancer cell proliferation have also been activated, so EGFR gene amplification kicks in as well in a number of patients. There are many other drivers. So you may need - some patients in the third-line savolitinib and Tagrisso alone will be fine.

But in others, you might need something else as well, something to address the EGFR gene amplification, which Tagrisso doesn't address. This is a chart that shows - it's a lot of data and it's probably something you need to study after this meeting.

But this just tolerability in non-small cell lung cancer in second line; and above, non-small cell lung cancer. So the top section of this chart shows all the approved therapies in second-line non-small cell lung cancer. So you've got Tagrisso, you've got Zykadia, you've got Cyramza, you've got the PDL-1 antibodies.

And what you see here is, looking at Tagrisso, the discontinuations on Tagrisso, it's about 13%, so it's very low. It's a very clean drug, Tagrisso. That's why it's doing so well. But then you go down and you look at Cyramza, which is Eli Lilly's VEGFR antibody, 37%. You look at Keytruda, the PDL-1 antibody, 37%. You look at next line down, also 36%.

So these are - again, these are monotherapies for the most part in that range of 20%, 30% discontinuation rates. You look in the second section here, these are the control arms, so this is the chemo doublet.

And you can see in the second-line setting chemotherapy, so the platinum and Alimta, pemetrexed, control arm in the AURA3 study from Tagrisso, 27% on chemo. Taxotere, docetaxel, alone 36% discontinuation rates, so quite higher. So then going down the savolitinib on the bottom section, you see, just for reference only.

You've got PRCC in that Phase II study, we have 14% discontinuation rate on savolitinib. So savolitinib monotherapy is a very clean and tolerable agent.

But when you put it in combination with Tagrisso, two very clean and tolerable agents, you end up with a combination discontinuation rate in the low 30s, which is, as I've said earlier, is common for first-generation monotherapies, let alone combinations. So this is not a bad place to be, and we're quite happy with that.

But this is why we're working, continuing to work on the dose of savolitinib and Tagrisso. And looking - this is 600 milligrams daily with no reduction based on the weight-based algorithm. This is just pure 600 milligrams. So the weight-based algorithm, hopefully, will bring that 30-odd percent down further.

And also, we're working on TATTON D which is 300 milligrams QD, which is a dose that we believe - well, we've seen efficacy, a lot of efficacy at 300 milligrams QD in the past in various of our dose escalation studies. So we'll see. We'll decide.

Once we've got the TATTON D data, once we've got mature TATTON B data, we'll decide what that chemo-doublet, controlled, randomized - global randomized study is going to go with.

Is it going to go with 600 milligrams with the weight-based algorithm? Or is it going to go with 300 milligrams? And we may start with both and then drop an arm as we get more data. Either way, ultimately, it's about minimizing discontinuations, maximizing tolerability and allowing these patients to stay on the drug as long as possible.

Because when they stay on this combination, they do really well. Safety profile, it's a chart we've shown before. Nothing really to say other than it's manageable and tolerable. Nothing that we worry about too much here. First line, we're moving on first-line non-small cell lung cancer, but we're keeping that fairly low profile at the moment.

And as time goes by, we'll talk more about that. PRCC, I won't go through it in a lot of detail, but this is the Phase II data we presented in 2017. I think you've all seen it, very strong PFS in this very difficult patients. 6-odd months median PFS in the MET-positive patients versus MET independent, just 1.4 months.

So this gave us a lot of confidence. Disease control rate, 73%. And then this is the molecular epidemiology data that basically shows in gastric cancer and in lung cancer, if you're MET-driven, you can see MET-driven disease is the red line, reaching median overall survival post-surgery in less than two years.

Whereas, if you're MET-independent, no MET-driven disease, you don't hit the median for 10 years. So MET is a really bad thing in the context of gastric cancer and in the context of lung cancer. This data doesn't yet exist in kidney cancer, and that's why we're doing the molecular epidemiology study.

If we can replicate discount the output with the PRCC molecular epidemiology study, which is much bigger than these studies, that will be very compelling data to take to the FDA to look for some kind of an accelerated approval for savolitinib in PRCC, MET-driven PRCC in our view.

Safety at the monotherapy in more detail, but basically, clean and tolerable. Gastric cancer, you've seen these charts before, really - I mean, we presented data, there are some updates in the announcement. We've screened over 850 gastric cancer patients in the last 1.5 years, 2 years.

About 5% of them are MET-driven and those 5% do really well on savolitinib. But that's an awful lot of screening to get a relatively small amount of patients. And these are very sick patients, actually later-line gastric cancer patients are very, very sick. So clearly, there's a signal. And clearly, there's opportunity there.

But our first - our primary objectives are lung cancer and papillary renal cell carcinoma and exploring more. And then once we figure out what are the highest potential additional indications, we'll figure that out once we've got our Phase IIIs enrolling in multiple indications in lung.

Fruquintinib, the Phase III data for FRESCO, you've all seen that, 9.3 months, median OS versus 6.5. Good efficacy relative to Stivarga, Bayer's third-line colorectal drug. The safety is where the big advantages for fruquintinib.

Deliver toxicity - hepatotoxicity for fruquintinib is really good in the context of regorafenib or Stivarga, where they have pretty significant challenges with regards to liver toxicity that they have to monitor closely.

And you see quite high levels of drug interruption, dose interruption and discontinuation down for Stivarga, 70% dose interruption because of monitoring liver function tests. And if liver function - if the enzyme, liver enzymes elevate, you have to take them off Stivarga. You don't have that same problem with fruquintinib, which is very clean.

FALUCA, the Phase II data was compelling in PFS. It wasn't powered for overall survivor, so it was - it looked good for the EGFR mutation-positive patient, but still not powered for OS, but powered for progression-free survival. So we'll soon see in third-line non-small cell lung cancer just how good fruquintinib is.

Big study we've got underway in China in first-line non-small lung cancer in combination with Iressa, 50 patients or so have now been treated and are progressing with their treatment, 76% response rate, safety profile looks good. Two small molecules in combination are really manageable because they have a short half-lives.

So fruquintinib, if you put it on top of an EGFR inhibitor, because of fruquintinib short half-life, if adverse events start emerging, you can remove fruquintinib. And within a day or two, it's out of the system. But if you put Avastin or Cyramza on top of an EGFR inhibitor, it a three or four week half-life.

So if AEs start emerging, you got to take them off of the EGFR inhibitor, and that's not good thing for EGFR-driven non-small cell lung cancer. So our view is small molecules are, in the context of combinations, are a much more practical way of treating patients with these combinations. And so we believe that this will play out.

And we're discussions with Lilly at the moment about expanding this program. The gastric cancer study with paclitaxel on fruquintinib, moving rapidly, and the Phase III is enrolling now. And I think, probably, we'll complete enrollment by the end of the year - sorry, not end of this year, end of next year.

Sulfatinib, I won't go it through it in a lot of detail. This is - this chart, Page 43, has been updated to show patients that respond to sulfatinib who have failed on Sutent and Afinitor. And there are 12 patients, 12 to 14 patients actually, who have failed on Afinitor and Sutent who then go on to do well on sulfatinib.

So that's what we're going to explore in America, is second-line post failure on those two approved therapies in pancreatic neuroendocrine tumors. And that Phase II in the States is - hopefully will kick off around the end of this year. A lot of preparation for that. Epitinib, I talked about. I want to show 70% response in brain mets.

You've seen these pictures before. This is new. The case study is an example of 120-milligram dose that I talked about earlier that doesn't have the pigmentation. Basically freckles is what it is. It's brown freckling as a sort of a cosmetic side effect. You can see here at 120 milligrams, great efficacy.

So we're not worried about going with 120 milligrams on fruquintinib, and we're keen to start that Phase III this year. I won't go through the others. 523, just for background, is moving rapidly now in both Australia and in China in hematological cancer patients. It's very exciting for us. It's been really hard work getting that going.

But now, we're starting preliminary efficacy, as you would expect. Syk inhibitor is upstream of BTK and PI3K delta, which are two targets that are really well-characterized in the context of hematological cancer. And ultimately, the mechanism of action, the scientific basis for it is very strong. The question is, is 523 a clean Syk inhibitor.

And it took Weiguo and the team 6, 7 years to design 523 in order to be a very clean and selective asset. So we're confident about that.

And actually, more importantly, down the road, it's going to be combinations in hematological cancer, particularly in aggressive types of non-Hodgkin's lymphoma like DLBCL, where you need to hit multiple targets to contain that disease.

In indolent, less aggressive, non-Hodgkin's lymphoma, you can get away with single agents like a BTK inhibitor or a PI3K delta inhibitor, or a Syk inhibitor for that matter. Patients still become resistant, so they'll have to have other treatments. And that's why a Syk inhibitor can become second line after a BTK inhibitor, for example.

But our view is, ultimately, combinations are going to be what matters. And having a really clean Syk inhibitor is something that is quite attractive for companies that are currently working on BTK inhibitors, for example.

Theliatinib, moving along in a esophageal cancer, Phase Ib/Phase II is enrolling in these EGFR gene-amplified patients, starting to see some nice results. And 689, the PI3K delta is, again, it's moving along.

So we'll start to be able to present some proof-of-concept data over the next year, 18 months on 689, which I think, our view is 689 is as clean a PI3K delta inhibitor as you've got, and it's the most selective one as well.

So you can see a chart down here, Zydelig, which is Gilead's PI3K delta, as very selective compound, but it's got a lot of toxicities around liver toxicity, primarily, and also - and colitis as well. The Bayer PI3K delta is also a PI3K alpha inhibitor, which can lead to some pretty significant infections as well.

So 689 is clean, it's selective and I think it has a great pharmacokinetic profile as well. So I think that's really going to make it quite differentiated versus what's out there. The FGFR inhibitor 453, also moving along and also encouraged by what we're seeing preliminarily in these studies.

I mean, ultimately, it's about picking patients with FGFR aberrations, and those patients do very well. So patients in bladder cancer, cholangiocarcinoma, they're most likely to do well. So that's the pipeline on the Commercial side. We just updated all of the charts, you can see all the market shares are growing rapidly.

She Xiang Bao Xin pill 15.4% market share, up from about 12% last year. So it's really motoring Bao Xin Wan. You can see the profitability, up 25% versus 2016. So very encouraging. Commercial organization continues to expand across all of China.

And here are some case studies of when we've taken over Seroquel, we've taken Concor, how we've been able to really rapidly grow those businesses over the last 2 or 3 years since taking them over from our partners. That's where I leave it now. Maybe open it up for questions. 9:44, so four minutes late. But here we go. [Operator Instructions].

So we'll answer questions in the room first, and if there are questions on the phone after that, we'll catch you later. So you have to have a microphone as well. Susie first..

Susie Jana, Edison. Good to see the development program for savolitinib in non-small cell lung cancer, but can you just make a little bit clearer now with so many different lines, what the sort of regulatory plan is? Should we expect submission of second line first? Just a bit of detail on that.

And on 523, I know that the FDA has requested some additional data. Could just give us some information on the nature of that data? And the last question is on the Commercial Platform business. That 10% to 15% net income growth is - show positive, and obviously, the top line has been distorted with a few things.

But should we expect similar top line growth rates if we take away all those impact as a historic? And where - how long do you think, if fruquintinib does launch mid to Q3 this year, how long will it take for sales to ramp up in China?.

Great. Thanks, Susie. There are all sorts of questions, I have to remember them..

I'll remind you if you forget..

So the first question is - it was non-small - it was - what's the regulatory plan on savolitinib. So to finalize Phase III pivotal registration studies, you need to go to the FDA and talk to them. You have to have year dose fully understood and laid out and agreed. That work is kind of underway right now.

I expect, as we say in our document, that those regulatory discussions will probably be around the end of this year, maybe even early next year. But AstraZeneca is being very aggressive. They're building this randomized, chemo-doublet controlled study, which is powered to be large enough to support ORR and PFS statistical powering.

We're doing that now. We're setting that up now. So that all gets led up to sort of the back end of this year when that can start. Hopefully, by that time, we've got the dose figured out. We can go to the regulatory authorities with everything, and then it can be converted from a Phase II to registration study. But that's to be determined.

So that's the plan on second line. Now third line is, it requires the same information, the same dose information. So third line, post Tagrisso failure, that information on dosing that plays out over this year will be used to set the strategy for registration in third line. The plan is fully to go ahead in third line.

It's just - we're not being as aggressive as we are in second line because, a, the signal is stronger in second line. You've got sort of 60% response rate in T790M-negative MET-positive patients on that combination. So very aggressive there. On the third line, it will come, it's just going to come probably 6 to 9 months later.

So that's the strategy, is you just go hard on both of those. The Iressa combination, as I've said, we're talking about China and savolitinib-Iressa combination, and that plays out this year as well. So I think that by the end of this year, we will have started the chemo-doublet global study.

We will have figured out, is it the weight-based algorithm 600 for the heavier patients and a lower dose for lighter patients, is that our dose? Or are we going with 300 milligrams for everybody? That's the question. We'll go with one of those.

We know that, worst-case scenario, 600 with the weight-based algorithm is going to give you something in the sort of 20%, around maybe 20%, 25% discontinuation rate, which isn't bad considering everything I've showed you today. So that plays out this year. First line, we're also working on that, and that will become more clear in as time goes by.

We have quite an aggressive plan on first line.

Okay, so does that answer your first question? Second question was, remind me?.

It was on the Syk inhibitor, what's the nature of the additional information that the FDA has requested?.

Oh, okay. So when we submitted our IND last year on the Syk inhibitor, the FDA asked us to go off and do three months tox testing on the M1 metabolite of the Syk inhibitor. So as 523 has metabolized, somewhere between 20% and 30% of the metabolites that come off of 523 is a single metabolite.

And so that's quite a big number, 20% to 30% of the metabolites being a single substance. So they said, look, have you got tox data on that? And we said, No. And they asked us to go back and run a three-month tox on it. So that three months tox is complete. The data is just about ready to be resubmitted to the U.S. FDA, there's no problems with it.

We expect the Syk inhibitor and the IND to be cleared, hopefully, before the middle of the year. That's that.

And then the last one?.

Just on the Commercial Platform, the top line sales and also just trying to put that in context that if you do launch fruquintinib this year, fingers crossed, how many years it might take to ramp up? Obviously, it's in the whole franchise..

So, yes. Okay. So Commercial Platform, the biggest problem with our Commercial Platform is that we don't consolidate the sales of our joint ventures, right, which are growing rapidly. And they're now, this year, $480 million or so, $470 million in sales. So we don't book those.

What we book is the sales of our 51%-held Sinopharm joint venture, Hutchison Sinopharm, which is affected by the Two-Invoice policy this year.

So there's a sort of a drop-off in sales, albeit no impact on the profit, right? I think once you take this step-off, because of the Two-Invoice policy, then you just sort of keep going at what we've seen in the past, I think in last couple of years, we've sort of said 10% to 15% growth top line and bottom line on our Commercial Platform is what we expect.

I mean, this year, the profit growth was 25% as we freed up, built all these new capacity and suddenly realized we've actually been pretty significantly capacity-constrained these past few years. But I expect on the Commercial side core business to just keep growing at 10% to 15%. I think that's conservative. I think we can do better than that.

In the context of fruquintinib, it's still going to be quite interesting because where - actually, where the Two-Invoice policy hurts us on Seroquel, it actually helps us on fruquintinib because we're the manufacturer.

So actually, Eli Lilly will not be able to book the sales of fruquintinib, we will, as the manufacturer, despite the fact that it's their commercial team that's detailing it. It's a bit messed up, but it is what it is. So as soon as we launch fruquintinib, we'll start to be able to report some of the sales or most of the sales on fruquintinib.

As far as how long it takes to reach peak sales, that's a difficult one. In China, actually, if you look at the innovative drugs that have been launched in China over the last sort of 10 years, particularly targeted therapies, they never reach peak sales, they just keep growing because there's such a big unmet medical need in China.

And it's a slow growth because you've got to get hospital bidding in all of the hospitals in every province to get into the hospital pharmacies. You've got to get on the regional reimbursement list and the national reimbursement list. It's a very gradual process of checking all these boxes, and eventually, you're everywhere.

A bit like where we are today on She Xiang Bao Xin pill, we're everywhere, fully reimbursed across all of China, and we're still growing. I mean, Bao Xin Wan was up 10% last year.

So I think - but to get to where we've said this sort of $110 million to $160 million in sales for third-line colorectal cancer, I imagine that's may be a five-year pathway to get to that number. And we say that's peak, but I think the market is going to be growing. And I think by the time we get to that, it'll go higher. Any other questions? Max..

Yes. Just to refresh - since I'm probably a little bit out of the loop, just on the Lilly fruquintinib.

Where are they ex China now?.

Yes. So when the FALUCA study reports top line results in Q4 of this year, that triggers Lilly's last option window, which is a two-month window. So just assume we report the results of the FALUCA study.

I don't know whether it'll be this stable, but say it's October 1, then they have all of October, all of November to decide do they want to exercise the global option and take the global rights. Now a few things are happening during that two months.

First of all, they're really studying the lung cancer data in the context of the FRESCO study, the third-line colorectal data, really looking at efficacy of fruquintinib, number one. But they're also taking into account the differentiation of fruquintinib in the context of Cyramza, their VEGFR 2 antibody, which is a pretty big drug for them globally.

But they've developed and they spent an awful lot of money developing Cyramza globally. So they've got some things to think about once they've got that lung cancer data. If they do decide to exercise the global option two months later, then they'll start a four-month period of negotiation of the global development plan.

Because they have to put forward full diligence, full commercially reasonable effort in developing fruquintinib globally if the exercise that option. And so that four-month period of agreeing what the global development plan is, is actually really important because we may not agree on the global development plan.

We may say actually that's not enough, you've got to go harder. And then they say, Oh, actually, that's what we can do. So that's how it plays out, two months for them to decide if they want it, then enter into four months of negotiation around what's the development plan going to be.

If we do agree with the development plan, then it gets done and then they take the global right.

And so that's the status basically, okay?.

Christian Glennie with Stifel. Just a bit more of sort of perspective and context, I guess, in terms of the dosing and - particularly on savo and the combination. Just in terms of whether this is something particularly new and obviously get the point about flex dosing and things.

But just something in terms of where it's come from in terms of your conversation with Astra, is it through regulators? And also, your level of confidence. And obviously, as you potentially step down on weight dose, you don't start to lose some efficacy from the drug..

Yes. This is very common in putting together your global registration plans, dose - and obviously, you have to have the efficacy, first and foremost.

If you've got no efficacy, there's no point in discussing dosage, right? Clearly, savolitinib and Tagrisso is a highly potent combination that delivers great benefit to a patient population that today has nothing, right? So that's sorted on response as well as the indications that I've given on duration of response and PFS, this is - put it aside.

So now it comes to saying, okay, do you want savolitinib to be $1 billion drug or do you want it to be a $4 billion drug? Well, for it to be a $1 billion drug, charge ahead with 600 milligrams QD, and you're going to have 30% of patients who's going to fall off the combination because of whatever reason.

Better, if you want patients to stay on it longer is to fine-tune that dose regimen and bring that number down.

Now you can see Sutent, with 36% discontinuation rate, is $1 billion monotherapy, right? But these days, I think to be competitive with the broad range of therapies that are now available, not in this particular indication, you have to doing better than Sutent did 15 years ago. So for us, it's a normal process for Astra.

It's a normal process of really understanding the data that's come out of the proof-of-concept studies and fine-tuning the registration trial strategy based on that data.

Now you've got to also remember, what we presented in October of last year was preliminary data, so patients continued on these regimen and we've continued to observe these patients. So you have more information as you go. But as I said earlier, I think this is - we're all really encouraged.

Actually, the word in the announcement encouraged by this data was not written by us, it was written by AstraZeneca. They are very encouraged by the savolitinib-Tagrisso combination. And it is squarely in their strategic area of focus, and we're moving as rapidly as we can. It's an exciting program.

So I think fine-tuning the dose in parallel with getting ready to start this global randomized chemo-doublet study that will be powered to get us to what we believe today, subject to FDA discussion to confirm that, but we believe this will be enough to get us to where we need to go, and that's all happening in parallel at the moment.

So it's exciting for us I think..

I just wanted to follow up because, obviously, things are moving at a good pace in terms of 523, 689 will be coming in to the clinic and some of the earlier-stage stuff.

Cash is strong at the moment, but are you looking to partner these in the near term? Or will you continue to invest? With R&D going up to $110 million to $120 million, it's going to go pretty quickly. So a little bit of guidance on partnering would be great..

Well, yes, R&D is going up to $110 million, $120 million. But you got the China business generating a lot of cash. You've got partners giving us a lot of cash as well. So the net loss for next year is $19 million to $52 million, right, somewhere in there. It'll probably end up in the $40 million somewhere, I imagine, maybe the high 40s.

So the cash we've got at the moment is plenty to see us through next year and well into the year. Probably 2020, I would imagine. I think on partnering, we're in no hurry.

We're actually in discussions on a bunch of stuff with regards to the Syk inhibitor, mainly in the context of combinations and kind of nonexclusive collaborations to look at how that compound does with other interesting compounds in the field of hematological cancer. So it's about building a data set, a proof-of-concept data set.

I think once we've built that up to being a deep data set, it will be an extremely valuable asset. Looking around, you've got Gilead, you've got to Takeda, they're the only two other players of note working on Syk. They're moving quite quickly, so we have to move quickly, too.

But I think Syk, over the next 12, 24 months as our data set emerges, we will be seen as really the only unpartnered Syk inhibitor that's far along. And so I think we would be high potential for big partners coming and looking to collaborate with us.

But we want to be in a strong position with a strong data set to be able to then decide if that's what we want to do or do it ourselves. We can decide later..

Just taking the view if Lilly - just going back to fruquintinib, if Lilly doesn't take the program, what's your global plans for fruquintinib?.

We're charging ahead. We're charging ahead. And actually, deep down, I kind of hope they don't because what we've got is - the deal that Lilly has, if they exercise the option, is a deal that we did back five years ago, when we were a much smaller company and it's actually quite a good deal.

I think one of the reasons that they wouldn't do it is because of the conflict with Cyramza. It's not because it's not a good deal, it's a great deal that they would get. So it's more that conflict with Cyramza.

And also, they'd have to be on the line for a massive global development plan, which would be a big investment for them on the back of having just done the same thing for Cyramza. And if there's an overlap, it maybe doesn't make any sense. So we are moving ahead as if they will not exercise their global option.

So the Phase I bridging study already started at the end of last year. We expect that by the time the FALUCA study reports, that will be complete, will be Phase II ready. Actually, we may even decide to identify some innovative combinations.

That's really what we're most interested, is fruquintinib in combination with other targeted therapies like savolitinib, for example. Fruquintinib and savolitinib in clear cell - or renal cell carcinoma, in our view, is a very exciting combination. It has great potential in combination.

So we might get to a point where late this year, we see Lilly is not going to exercise, we finish our Phase I bridging study and we're ready to just get quite aggressive in some combinations. And that leads to another thing that we are doing at the moment, which is starting to build up our clinical regulatory organization in the United States.

We've - historically, we've run all global development under our control from China working with CROs and third parties around the world, we've concluded now that we actually have to put our organization on the ground in the U.S. to be in readiness for this.

So as sulfatinib goes into Phase II, as fruquintinib starts to potentially get into some interesting global development and U.S. development, we want people on the ground to run that for us..

Any other questions? Okay. All right. So if there is no more questions, 10:06, sorry we're a bit late, but thank you all very much for coming and for your interest. Thank you..