When Tony Wood became GSK’s top scientist in mid-2022, the British pharma was waging battle.
GSK was fighting Zantac-related litigation in court while struggling to calm investor jitters as its stock sank. Though CEO Emma Walmsley triumphed over activist investing giant Elliott Management’s argument for her ouster, bigger questions lingered about whether GSK had the right pipeline strategy.
Earlier this month, Wood passed his two-year mark as chief scientific officer. GSK is on firmer footing. Its stock is up about one-third from its nadir in 2022, driven by the company’s shingles and RSV vaccine blockbusters. And despite several setbacks, even its oncology pipeline has some momentum.
In a wide-ranging conversation this month, Wood talked with Endpoints News about his vision for GSK’s research, from cancer plans to the obesity market and AI’s impact. This conversation has been substantially edited and condensed for length and clarity.
Andrew Dunn: GSK went from exiting cancer with the Novartis swap in 2015 to reentering under Walmsley and then-CSO Hal Barron. Would you call GSK a top pharma in cancer?
Wood: I’m really pleased with the progress the portfolio is making. Oncology is emerging but in a careful way.
We’re delighted with Ojjaara as a really nice piece of business development. It’s making a fantastic start as a launched medicine in the myelofibrosis population.
I didn’t believe for a minute we’d get through this conversation without talking about Blenrep. [Editor’s note: After winning an accelerated approval in 2020, Blenrep was removed from the US market in late 2022 after a negative Phase 3 readout. GSK is now in the process of applying once again for approval, with more Phase 3 results.] We’ve been very careful to not include Blenrep in any of our revised growth projections. That’s simply because launching a new medicine is a very different proposition than bringing one back from having been withdrawn from the market.
The ocular side effects that were initially viewed in DREAMM-3 as being an impediment for widespread use are manageable and reversible. We’re in good shape on that. The submission was accepted in Europe just recently, and we’re getting close to the submission in the US.
In women’s cancers, we have the emerging position that Jemperli is establishing, particularly in endometrial cancer. We got approval from the FDA in all comers there. We’re the only ones with an OS [overall survival] label. Keytruda doesn’t have an OS label.
That is the bridge into other solid tumors, potentially lung and colorectal. One has to walk into that highly competitive area with eyes wide open. I’m very data-driven. I know you speak to Hesham [Ahmed Abdullah, head of oncology R&D] a lot. What I need to see from him are results that show we’re going to be different and therefore potentially competitive with a major market leader rather than just a participant.
I leave TIGIT right to the end. That’s really the poster child of, are we going to be different from the larger players or not? We’ve got early results, which hint that we might be. But I need to see more before I would declare we have a path forward that is going to be different from Roche.
Dunn: TIGIT may be immuno-oncology’s most debated and polarizing target these last few years. Why did you decide to enter Phase 3 when Roche, Bristol Myers and Novartis are walking away? Why are these other big drugmakers wrong and GSK is right in staying with TIGIT?
Wood: I wouldn’t quite characterize it in such binary terms at the moment.
We’re cognizant of what happened between CITYSCAPE and SKYSCRAPER and the diminution in signal there. What’s different about the data we have in front of us? We’ve got a larger group from the platform study, nearly 100 patients, and then you look at the speed and depth of response. It looks different. We haven’t published any of that. It’s going to come at the end of the year.
Although we’ve started the Phase 3 study, it’s very heavily gated. We’ll continue to test if the apparent difference — which is not in overall response rate, but in depth and speed of response — is consistently building as we see more patients.
Why might it be different? I’ll give you two reasons. SKYSCRAPER had a less effective PD-1 regimen. We have in Jemperli a more effective PD-1, I feel, particularly in the high PD-1 expressing population where you see the biggest difference.
Lastly, we have a TIGIT molecule that has a modified functional region. That modification is consistent with cytological changes that you would expect to see in the tumor.
We’ve got a range of different features, which all add up in a cogent way to make me feel we might have something different. But I’m still being cautious about testing whether or not I have that as I go through the Phase 3 study.
Dunn: I constantly hear about the challenges of bureaucracy, like size versus nimbleness. In three decades in the industry at Pfizer and GSK, what do you think is pharma’s R&D core competency?
Wood: You’ve got to look back at what pharma was doing in the mid-1980s onwards when it was truly successful.
It took two disciplines: organic chemistry, which had turned into medicinal chemistry; and biology, which turned into the basis of drug discovery. And it scaled them in a way you couldn’t find in universities or in biotech. You get this advantage of scale and experience, and from that, you learn what works and what doesn’t. At the time pharma was lucky. Biology just happened to translate.
Big Pharma is really good at building scaled, effective, and quick solutions. That’s one of the first things I did when I first started at GSK in 2017, which was to shift them from being heavily focused on inhalation and building out that broader platform.
Dunn: Under your CSO tenure, GSK bought Bellus Health for $2 billion and Aiolos Bio for $1 billion. Why haven’t we seen bigger deals?
Wood: My attitude to large-scale mergers and acquisitions is just that it’s hardly ever worked in the history of the industry. Where it has worked, it’s been the acquisition of a technology at a point in time when that technology is just maturing, such as monoclonal antibodies with Genentech/Roche.
We tend to focus on bolt-ons. Aiolos was consistent with our long-acting portfolio in respiratory. That was doing two things for me: It was adding an opportunity to treat patients with so-called low T2, and I needed a long-acting TSLP [or thymic stromal lymphopoietin].
I’m always looking for things where I can see a way of doing a better job of the development than the original sponsor. TSLP, we learned from depe. [That’s short for depemokimab, an IL-5 antibody]. We were able to go straight from Phase 1 to Phase 3 — all parallel indications — in four years versus seven in the case of Nucala.
Dunn: Aiolos’ leaders expected to start Phase 2 in asthma in 2024, and GSK called it a Phase 2-ready drug when it acquired the company. On the latest earnings call, you said you expected to start Phase 2 in 2025. Why the delay?
Wood: It’s about optimizing, so I can go quickly from Phase 2 to 3.
Dunn: So you’re saying you can apply some of the ideas of going Phase 1 to 3?
Wood: That’s right. In the case of depe, we had a very good PK/PD understanding coupled with the appropriate way of measuring potency from Nucala, which allowed us to essentially do a Phase 1b and model out what the Phase 3 efficacy would look like.
We’ve got a little more detail to collect around TSLP in order to do that. What I want to do is to get that bit right so we can go quickly into Phase 3 and a broader set of indications. The objective here is to go quicker than was originally anticipated.
Dunn: AI is having a moment in areas like structural biology. But is it moving the needle today for tasks like how trials are run, clinical decisions are made, and regulatory packages are put together?
Wood: There’s no doubt what’s going on in the small molecule world and mostly the protein work coming out of David Baker’s lab is having a big impact on creating a more effective way of designing molecules. But I would say that’s in many ways polishing diamonds. We’re pretty good at that anyway. And if I solve that problem, I’m not solving the real problem, which is that most things fail in Phase 2, going back to targets and patients.
In development, we’re already making good progress with, let’s call it, the more prosaic aspects you mentioned. Automated filings, things like that. All of that needs to be worked through carefully with the agency.
The thing I’m really excited about, that I think will come in the next five years certainly, is a much more effective way of designing and recruiting into Phase 3 studies, so we can identify patients for whom we can see effects earlier.
We mentioned Nucala. If you have an eosinophil count of more than 300, you’re going to respond. I would like that to be a feature of every one of our medicines: We know who’s going to respond.
The data on patients is getting better and better. We need just to fill in those last gaps. Then we move from a world in which the pharma industry is collectively successful, but you get random individual companies enjoying that success at any period of time, like PD-1, now GLP-1. Into one that, if it’s done properly, we’ll be able to see a greater degree of sustainable success.
Dunn: I’d be remiss to not ask about obesity, which is obviously absent from GSK’s story. Do you have any interest in starting research on obesity, given the size of the market and magnitude of benefit of GLP-1s?
Wood: No. And for two reasons. I have enough opportunity in front of me with what I have in the portfolio already. And I’m judging that future opportunity through a lens of what the world will look like after the GLP-1s have the effect they have on body weight as a comorbid factor.
Dunn: Say more on that. What do you see as second-order effects?
Woods: Take an area like MASH. There will be a legitimate comorbid effect associated with reducing body weight. If you look at the more fibrotic end of the spectrum, we’re not predicting body weight reduction will have the impact there that it does in the earlier phases of disease.
I don’t want to go into too much detail here, because I’m going into territory that is not disclosed, but neurodegeneration. We’re very interested in the recent results that you see from shingles and Shingrix vaccination. Although there are claims that GLP-1s reduce neurodegeneration, it’s probably vascular rather than more broad in nature. All that’s going to do is push the curve further rightward.
We all have to be cognizant that in 10 years’ time, we’re dealing with a world that is the consequence of having controlled body weight to the degree that GLP-1s and the follow-on agents from that will control it.
Dunn: It usually takes at least a decade to see an R&D strategy play out. What would you like to accomplish by the end of your tenure?
Wood: I’ve got a very clear view of what I think needs to happen in order to create a more sustainable, productive base that will not only fuel the growth of GSK. I’m confident that I’ve got the portfolio and the technologies to do it but hopefully will set a path for how we’ll do it more broadly in the industry.
I’m lucky enough to have discovered medicines in my career. I want to really put this thing on a more sustainable footing.