Investors "get it" that Gilead (NASDAQ:GILD) has two best-in-class antiviral franchises, HCV and HIV/AIDS. However, with some reason, the Street is skeptical that GILD can come close to reproducing this degree of success outside of antivirals with an acceptable ROIC.
To evaluate this I begin an occasional series of articles evaluating the promise and perils of some key parts of the company's pipeline.
GILD is currently testing more than one drug for more than one seemingly unrelated disease. Of these, the one that is under consideration for the most diseases and that is farthest along in clinical trials is an antibody called GS-5745. I believe this came along with simtuzumab in 2011 as part of the $225 MM acquisition of Arresto, a privately-held biotech company. GS-5745 is a humanized antibody; the humanization process was performed for Arresto by Antitope, a British company that is now part of Abzena plc (ABZA.L). Abzena is said to be in line for very small royalties on simtuzumab and GS-5745.
Both simtuzumab and '5745' were developed to inhibit enzymes that act outside of cells, such as in the extracellular matrix. Simtuzumab inhibits an enzyme involved in collagen (fiber) formation. It is in Phase 2 for fibrosing stages of the liver disease NASH but failed in clinical trials for cancer and pulmonary fibrosis. GILD's CSO, Dr. Bischofberger, recently admitted that these failures make the drug's chances of success in NASH problematic. None of this was surprising to me; if you go back over my many GILD articles, I've never had anything especially optimistic to say about simtuzumab.
Does this reflect badly on the other Arresto drug, 5745?
Probably not, but the risky nature of drug development is shown here, especially for a company such as GILD that is generally intent on doing original research rather than me-too drugs.
5745 may be important from a trading standpoint for GILD, though not in the short term. At a closing price Tuesday of $81.79 and a TTM GAAP P/E just under 7X, I think that at some point, GILD will likely begin to reverse its downtrend if it can give new investors clear evidence that it can bring innovate outside of antivirals.
My view remains that it's not GILD's absolute P/E that's the problem, given the belief that earnings are headed down, it's the relative P/E. With projected 2016 sales of $30 B or more, GILD is approaching the size of Big Pharma. Essentially, every one of those companies has proven that it has been unable to grow sales and earnings for years. Yet their P/Es using GAAP are typically in the 25-35X range.
All these big companies are going after the same targets, more or less. Thus they all are subject to the same competitive forces that pressure sales and margins.
And the market as a whole (NYSEARCA:SPY) is trading at 24X TTM GAAP P/E, with EPS showing a CAGR over the past 28 years below 5% - barely above the rate of inflation plus population growth. Anyway, I'm sticking patiently with my GILD allocation, and keeping on top of its pipeline to see if and when it really can deliver innovation. The company organizes its pipeline web page by therapeutic category, which is how marketers do it. However, the way scientists do it is think of molecular mechanisms and then see which areas of unmet medical need could be served by modifying those molecular processes with a compound or combination of compounds. We call these compounds 'drugs.'
Given the hopes for MMP inhibition in cancer and other diseases, this particular pipeline review follows the latter approach and discusses the potential for success, and commercial and financial relevance, for 5745 in certain inflammatory diseases and cancer.
Except for some comments on the stock I may add, this article and series are more for serious pharma and GILD investors or people simply interested in the R&D process or a specific disease.
I've done my best to make this as accurate and accessible to general investors as possible. If any readers have technical knowledge and find an error or have a materially different point of view, please comment.
First, a brief overview of the general rationale for GILD to study 5745 in several diseases.
What GS-5745 may be able to do for GILD
This antibody addresses a field that has attracted most Big Pharma companies about 20 years ago. In that period, a group of extracellular enzymes called proteinases was felt to have important roles in various disease states, and small molecule inhibitors were developed. These MMPs (matrix metalloproteinases) comprise a group of over 20 distinct proteins (enzymes or proenzymes). Most of them reside outside of cells; some are attached to the cell membrane. The different MMPs are simply identified by number, MMP-1, MMP-2, etc.
GS-5745 is aimed to inactivate MMP-9, also called MMP9, also called gelatinase B, which is elevated in a number of diseases.
Since the 90s, more basic science knowledge of the role in MMPs in a number of diseases has been gained, which may make GILD's task here just a bit easier than it was when MMP R&D was going hot and heavy.
This sector proved to be a graveyard for drug development. A 2015 article that relates to one of the lead targets for 5745, ulcerative colitis, has an abstract that while somewhat technical, describes a part of the history of this effort, why it failed, and why 5745 could be a breakthrough drug for GILD (bolded emphasis added):
Expression of matrix metalloproteinase 9 (MMP9) is elevated in a variety of inflammatory and oncology indications, including ulcerative colitis and colorectal cancer. MMP9 is a downstream effector and an upstream mediator of pathways involved in growth and inflammation, and has long been viewed as a promising therapeutic target. However, previous efforts to target matrix metalloproteinases (MMPs), including MMP9, have utilized broad-spectrum or semi-selective inhibitors. While some of these drugs showed signs of efficacy in patients, all MMP-targeted inhibitors have been hampered by dose-limiting toxicity or insufficient clinical benefit, likely due to their lack of specificity. Here, we show that selective inhibition of MMP9 did not induce musculoskeletal syndrome (a characteristic toxicity of pan-MMP inhibitors) in a rat model, but did reduce disease severity in a dextran sodium sulfate-induced mouse model of ulcerative colitis. We also found that MMP9 inhibition decreased tumor growth and metastases incidence in a surgical orthotopic xenograft model of colorectal carcinoma, and that inhibition of either tumor- or stroma-derived MMP9 was sufficient to reduce primary tumor growth. Collectively, these data suggest that selective MMP9 inhibition is a promising therapeutic strategy for treatment of inflammatory and oncology indications in which MMP9 is upregulated and is associated with disease pathology, such as ulcerative colitis and colorectal cancer. In addition, we report the development of a potent and highly selective allosteric MMP9 inhibitor, the humanized monoclonal antibody GS-5745, which can be used to evaluate the therapeutic potential of MMP9 inhibition in patients.
All authors of the above were GILD employees. So this is a sort of mission statement and raison d'etre for the company's focus on this antibody. Basically the company is hopeful that the research that failed was well-directed but had too unsafe a set of drug candidates; and the MMP9 is an adequate target for at least one commercially important disease.
The above lays out a risk here: GILD is doing pioneering work. Every effort to modify MMPs for therapeutic benefit has failed; MMP9 has both good and bad effects in cancer; so, inhibiting its action may have unpredictable effects. The bulk of the specific diseases that GILD is testing 5745 for are different cancers.
However, I'd like to start with the clinical program in inflammatory bowel disease, first UC and then Crohn's; then in COPD; then in various malignancies. Put together, these are very large efforts, and to my knowledge, GILD is out there alone among major companies with this antibody approach.
GS-5745 for ulcerative colitis
GILD has been studying 5745 in both pre-clinical models of and in people with moderate-to-severe ulcerative colitis. This is a non-orphan niche market that remains underserved despite a growing number of approved treatments, with more apparently on the way. GILD itself has an oral agent, filgotinib, in or entering late-stage development for Crohn's.
UC is a serious inflammatory illness that predisposes to colon cancer. The article quoted from above discussed pre-clinical data on UC and colon cancer, and was felt promising, leading to a Phase 1 study.
GILD has now published Phase 1 results (this link is to the summary, which in turn provides links to the full article). The conclusion of the summary is:
This phase 1 trial provides preliminary evidence for the safety and therapeutic potential of GS-5745 in the treatment of ulcerative colitis.
The company judged safety to be satisfactory. Details of the responses were as follows:
Clinical response occurred in 18/42 patients (43%) receiving GS-5745 compared with 1/8 patients (13%) receiving placebo. Clinical remission occurred in 6/42 patients (14%) receiving GS-5745 and 0/8 (0%) receiving placebo.
Since the sample sizes were small, there could be no statistically significant differentiation between test and placebo, but the company liked the trend. (Some readers might be interested in a more detailed review of the same study, focusing on the histology.)
As is the case routinely with so many studies now, this was really a Phase 1 study with Phase 2a characteristics. So there was evidence of concept, but no proof of it. Nonetheless, as is the case with much but not all of GILD's 5745 efforts, the company is eager and has moved this to a registrational, Phase 2/3 study.
This is a multi-year study with an estimated enrollment of 1600 patients. The study began in September last year; completion date is estimated in Q1 2019. As I understand the study design, the first part of the study is an 8-week study looking for remission rates with 5745 given as 150 mg subcu either once a week or once every two weeks.
This is "cohort 1" and is the Phase 2 part of the study, which I assume is ongoing now.
Then if satisfied, a 52-week remission study will be performed. The ClinicalTrials.gov summary of the study design is not crystal clear on exactly how and on whom the Phase 3 study would occur. But that's a detail that investors may not need to think about.
I hope that the company would report on the 8-week part of the study at an appropriate time.
If approved, say in 2020, 5745 will face an increasingly competitive UC market, so there are significant limits on sales potential. I'm mildly optimistic that this drug for this indication can succeed for this indication. If the 8-week data show no efficacy, or if there are safety issues, based on past experience there may or may not be no press release or other mention of the study's early termination. This listing may simply disappear from the pipeline web page.
Thus the longer this indication stays active on the website, the greater the chance that 8-week data were good enough to go to Phase 3.
GS-5745 for Crohn's disease
The other major autoimmune inflammatory bowel disease, Crohn's disease, has some published evidence that MMP9 inhibition might be helpful. Blood levels of MMP9 have been found to be one of a group of biomarkers for Crohn's.
Some clinical evidence from infliximab (Remicade), which is approved for Crohn's, also may support the benefit of 5745 in this disease. From a 2015 review article of MMPs and IBD, with comments on several different studies:
The effects of the clinically used infliximab on gelatinase levels in CD patients were examined and found that MMP-9 serum levels were consistently decreased following infliximab treatments and MMP-9 expressing polymorphonuclear leukocytes were also reduced in biopsy samples... Using mucosal explants of IBD and control patients, infliximab was shown to downregulate MMP-1, -3, and -9 levels... Most MMP expression was increased in IBD patients with the exception of MMP-28 which was downregulated and responders to infliximab had a gene expression and gelatinase activity [MMP9 = gelatinase B] was restored to control levels following treatment.
So far, so good in the article. The authors caution extensively as follows in their conclusion, though:
However, the potential to therapeutically exploit these discoveries raises several questions. Paramount to the discussion of MMP inhibition is the potential knock on effect of the inhibition. Blocking a single enzyme will affect activation of other proteases and its considerable list of substrates which is likely to disrupt physiological processes...
Consistent with the above caution, apparently since less is believed to be known about MMP9 and Crohn's, GILD has followed a less aggressive strategy with 5745 for this disease. It has begun a Phase 2, 8-week study in about 175 patients with moderate-to-severe Crohn's. Results may be reported to the public in Q1 2017.
Interim summary: At moderate expense, GILD is exploring a promising therapeutic alternative for cases of inflammatory bowel disease. Given that it wants to be in the IBD space with filgotinib, it's following industry norms in trying to build a marketing franchise. Success in both indications will not "make" GILD, but every little bit helps, and perhaps good news in one or both of the Phase 2 IBD studies may buoy investor sentiment toward the stock.
Before discussing GILD's opportunity with 5745 in cancer, some comments about its program in COPD.
GILD tests 5745 in cystic fibrosis
While the GILD pipeline webpage lists COPD as a target disease, the specific form of COPD that is being studied in Phase 2 is in CF.
The title of the study is Effect of GS-5745 on FEV1 in Adults With Cystic Fibrosis. FEV1 is a measure of lung function, more specifically ability to move air when exhaling.
GILD markets Cayston for CF, so it knows the disease well. In this, lungs get clogged with mucus. I've done a little reading on this topic. It would appear that at the very least, MMP9 is a biomarker for CF activity. Whether modifying this will be safe and effective, and possibly effective for any other complications associated with CF such as pancreatic problems, is to be determined.
While this study is listed as a Phase 2 trial, its design has similarities to the Phase 2/3 UC study. This raises the question in my mind as to whether GILD might treat this study as registrational if the results are strong enough.
A recent review titled Matrix metalloproteinases in destructive lung disease, and other data I have looked at, raise serious questions in my mind as to whether this target is a long shot.
Success in CF could open up the broader field of COPD or more specific lung diseases to study by GILD; but I'd give this indication no value right now.
Interim summary: The use of 5745 for CF or other forms of COPD appears speculative to me, but the current Phase 2 program in CF will not cost GILD much. Thus I don't look at it as a mistake, just one of those things that large companies do either as long shots that might pay off, or just to guide further research.
Now, a discussion of GILD's varied programs for 5745 in cancer.
Introduction to MMP9 and cancer
A large number of malignancies overexpress MMP9. This has, for at least 20 years, made MMP9 or MMPs, attractive potential targets for anti-cancer therapy. With what may prove to be the first safe "magic bullet" for MMP9 in pharma history, GILD is doing pioneering work in this field. Unfortunately, it has no head of oncology, and the role of MMP9 in different cancers is not straightforward. MMP9 has been examined in a number of studies; I've read some reviews of the subject, but have no research expertise in it independently.
My understanding is that 5745 could be beneficial, or could be harmful, and that it could depend on the stage of a certain cancer, or exactly what other drugs it is given with. In other words, so long as high quality institutional review boards are comfortable with the studies, then if they do not cost a lot, GILD can have a positive risk-adjusted ROIC by studying 5745 in cancers even if the chance of success of any one study is below 50%, given the durable upside.
That's how I look at this program: risky, but something worth doing; and a great head of oncology would help.
First, the Phase 3 study of 5745 in gastric cancer; then, a word on some hope it can work in pancreatic cancer; then some comments on the many Phase 1 (really, Phase 1/2) studies in several other types of cancer.
Rationale for and study of GS-5745 in gastric cancer
At the ASCO meeting this year, GILD had an abstract presented that reviewed the clinical and safety data of its Phase 1 study. Some quotes from the abstract that provide background information and are interpreted by GILD as justifying further, larger-scale study:
Background: GS-5745 is a monoclonal antibody that inhibits matrix metalloproteinase 9 (MMP9), an extracellular enzyme involved in matrix remodeling, tumor growth, and metastasis. Preclinical studies demonstrate that MMP9 inhibition alters the tumor microenvironment, associated with greater chemotherapy penetration and improved anti-tumor immunity...
Results: As of January 12, 2016, 40 patients (pts) with metastatic GC were treated (14 remain on study)...
Median progression free survival (NYSE:PFS) for all 40 pts is 7.4 months... with a median duration of response (DOR) of 9.4 months and an objective response rate (ORR) of 50%. Of those pts, 30 pts were chemotherapy naïve (eg. 1stline), and demonstrated a median PFS of 12... months, with a median DOR of 11 months and an ORR of 57%. Collagen neoepitope decreased with continued treatment, demonstrating on-target effects...
Conclusions: Adding GS-5745 to mFOLFOX [5-FU plus a platinum-based drug with leucovorin] is feasible and well tolerated, and associated with encouraging activity. Biomarker studies demonstrate MMP-9 inhibition. For treatment naïve pts, the median PFS was considerably longer than expected (12 months). GS-5745 is now being examined in a registrational phase III study in 1st line GC with mFOLFOX chemotherapy.
This is encouraging data but not proof of safety and efficacy. Thus GILD has appropriately begun a registrational study, which was also the subject of an abstract at ASCO. Science nerds note the difficulties here; per the abstract:
The study is designed to have an 85% power to detect clinically meaningful improvement in overall survival at the one-sided significance level of 0.025.
This means that even if the drug works with a reasonable P value, there's also a reasonable chance the data will not show it.
This is a 430 patient study that is estimated to end September 2018. As opposed to the CF and IBD indications, 5745 is being given with chemo, at a high dose intravenously.
The market size for gastric cancer is in the billions of dollars. In the US, the disease was substantially more common 80 years ago than it is today.
Next, a brief discussion of 5745 and pancreatic cancer, which has also been investigated in Phase 1.
Some hope for 5745 in cancer of the pancreas
Also for this year's ASCO, the following title was not presented but was published in association with the meeting:
Results of a phase I study of GS-5745 in combination with gemcitabine and nab-paclitaxel in patients (pts) with advanced pancreatic cancer
The results were indeterminate, with the abstract ending with this conclusion:
Adding GS-5745 to GA [gemcitabine/Abraxane] does not appear to add additional toxicities beyond those seen with GA therapy alone. In this small group of 31 pts, the addition of this MMP9 antibody may improve the median PFS and ORR over standard of care in previously untreated pts. Further studies are needed to investigate these potential beneficial effects.
Additional Phase 1/2 study in this disease is ongoing as part of a program in multiple other solid cancers.
Other cancers are under investigation, as well
As part of its large exploration of MMP9 inhibition in cancer, GILD is studying 5745 in other cancers. These include (see above link):
- non-small cell lung cancer
- colorectal cancer
- breast cancer.
These are safety studies using different chemo regimens as is appropriate for each disease plus 5745. Estimated date of completion is February 2018.
While I've ended with this study, obviously this is the "biggie" with this drug.
What about RA?
Last year, GILD completed a Phase 1 study of 5745 in rheumatoid arthritis. It continues to list this as a Phase 1 program on its website, but there is no mention of a Phase 2 study on ClinicalTrials.gov.
I'm going to ignore this crowded indication for 5745 for now and assume the company is deferring any clinical investigation of 5745 in this disease for now.
Concluding thoughts - will 5745 make a difference for GILD investors?
Success with 5745 in cancer is fairly 'iffy' as I see it, but the early data on gastric cancer means that I'm taking this effort much more seriously than I took the simtuzumab studies in cancer. Cancer is so complex, and the roles of 5745 in cancer so varied, that I'm looking at this effort as one with a modest chance of a big success, but one that is not being focused on by the Street. Given the moderate and not overly-expensive effort GILD is putting into 5745 and cancer, overall I like the program; albeit a full-time top-tier oncology chief would help me like it more.
In IBD, where filgotinib appears to have solid prospects, the company has put forth what appears to be reasonable evidence that it can succeed in UC and maybe also in Crohn's. If GILD can build an IBD franchise, it can add to its profits and provide another vehicle for further growth later.
Finally, the CF effort is worth following but commercially has to be small. If lives of CF patients can be helped, though, the effort is worth it on that basis alone; and perhaps some insights into more common inflammatory/obstructive lung diseases will be gained.
Putting this effort with 5745 in perspective, it is occurring in a company that is trying to make the transition from a heavily antiviral company to more of a traditional large company - but one focused on best-in-class and cutting-edge larger niches. I think this is the sort of effort that is called for, given the varied upside scenarios. The company is not risking much financially with these studies, will at the least obtain proprietary knowledge it may use in the future, and may enhance its image among investors and its business prospects if it can overcome the MMP curse and bring 5745 to market for at least one indication.
Thanks for reading. Any comments, and any information about this topic, are welcomed.
Disclosure: I am/we are long GILD.
Additional disclosure: Not investment advice. I am not an investment adviser.