Category Archives: AbbVie

Biogen Idec, multiple sclerosis, and the anti-Lingo story

It’s AAN conference week, and we were looking around, trying to get caught up on multiple sclerosis after a few months dedicated to oncology. We stumbled across this analyst report, and just had to comment.

Credit Suisse (CS) recently released a deep-dive report on Biogen Idec’s (Nasdaq: BIIB) anti-Lingo antibody program, assigning between $5-10BB (billion) USD of the total relapsing/remitting Multiple Sclerosis (rrMS) market share to the program by 2020. The program is currently in Phase 2. This analysis, in part, supports CS’s current price target for BIIB stock at $400, leveraging presumed growth due to the view on continued success of the anti-Lingo program. In other words, positive news on this program will help support inflated multiples through 2018, when pivotal trials may actually read out. The analysis seems ill considered, misses key aspects of BIIB corporate strategy, and places undo pressure on an early Phase 2 program. Further, intense focus on the anti-Lingo antibody program in turn places pressure on two early phase clinical trials, one due to read out in 2H14. The implication is that the base case for the price target could be undermined if the very early clinical development of the anti-Lingo program falters. That’s an unfair burden for a single high-risk program to bear.

Let’s dive in. Our focus will be on the science, but we’ll first set the stage. Our driving goal when looking at any biotech company or program is to bet the science, not the hype.

Two years ago the company set an internal goal of “400 in 5”, essentially promising to drive EPS in support of a sustained stock price of $400 USD by 2017. They came close during the biotech bubble that burst earlier this year. The stock is holding its’ own at around $320 USD. The “400 in 5” goal is in place irrespective of the success or failure of the anti-Lingo program, which cannot read out pivotal clinical trials until at least mid-2018. With that in mind we can deconstruct the CS analysis, and create our own. Importantly, our analysis drastically de-risks the impact of the anti-Lingo program on the trajectory of BIIB growth, while leaving room for very attractive upside if this program hits.

The CS analysis correctly estimates that oral MS drugs will take over an increasing % of market share running from 2014 through 2020. No argument there, and BIIB will take the bulk of this market with Tecfidera, per multiple analysts. But CS believes that the “pipeline focus” is on the anti-Lingo antibody program to will help drive the stock price as the program matures. A few comments:

1) Analyst and/or investor focus on the anti-Lingo program is a sign of pipeline weakness, not strength. Where, one might ask, is the rest of the pipeline?

2) The program is very high-risk (and thus high return) for multiple reasons beyond the inherent weakness of being in Phase 2.

3) Management recognizes the oversized risk of the program, and will not tether stock performance to this program, instead they will act to de-risk the pipeline and performance.

Let’s look at these points one by one. First, the portfolio and pipeline. We agree that top-line growth will continue to be robust, driven by Tecfidera in the expanding orals segment of the market. We believe that the Daclizumab program is likely to succeed (the data being shown at AAN this week is very good) but it seems likely that this drug will compete for the declining injectable biologics market share with Tysabri. Maybe not, if it is successfully positioned for JC virus antibody positive patients, and can hold off the orals. Ocrelizumab may successfully evolve into the successor for Rituxan, an anti-CD20 antibody pulled from the MS market by Genentech/Roche because of exposure to generics competition. STX-100, an excellent program for fibrosis, is emerging into a rapidly evolving IPF treatment landscape (pirfenidone, nintedanib), and we’ll see if the company can eventually steer this drug into other indications, such as systemic sclerosis. The hemophilia biologics Eloctate and Alprolix are approved and launch-ready, with a consensus view that these will pull in 500MM over the first full year of sales, rising to 1BB by 2018. That’s already baked into the current forecasts.  The rest of the programs are as high risk as the anti-Lingo program, so let’s be conservative and assume half or more of these programs eventually fail. Point #2 is that the anti-Lingo program is high risk and can fail for a variety of reasons. At least three can be articulated. First, the therapeutic hypothesis, that axonal regeneration can be induced by a therapeutic in the setting of MS, has never been demonstrated. So there is an inherent biology risk. Second, the preclinical data package supports the hypothesis that blocking Lingo will improve myelin sheath regeneration and axonal function after insult or injury. However the preclinical package using MS animal models is very weak. Finally, the technical hypothesis, that sufficient quantity of antibody can be delivered across the blood-brain barrier in a robust and reproducible manner, patient to patient, has not been demonstrated. So there is an inherent technical risk. It’s also critical to note that the optic neuritis trial, the first Phase 2 to read out, perhaps addresses the therapeutic hypothesis (we could debate this, but won’t) but simply fails to address the technical hypothesis. Focusing investor attention on a Phase 2 readout in optic neuritis as a surrogate for efficacy in MS is a shell game that will go bad quickly if that Phase 2 trials comes in with negative results.

So we agree with CS that anti-Lingo antibody might work in rrMS, and it might not. We disagree that this program should be the focus of interest in the pipeline. We disagree outright with a few of their more outlandish predictions, including the statement that anti-Lingo “has potential in SPMS” the progressive and untreated form of the disease. There is no support for this statement. And while we agree that anti-Lingo is likely to be used in combination with other BIIB MS drugs, trials supporting such use are a very long way away. There is no basis to evaluate such a statement at this time. Finally, instead of concluding that pipeline focus on the anti-Lingo program is a positive, as CS does, we see this as a sign of a fundamentally weak BIIB pipeline.

Should we be surprised? Let’s consider that BIIB has not successfully developed a novel internal program since Avonex and Amevive, well over 15 years ago (yes there is Peligry, but that’s just pegylated-interferon, still, they did develop it). What else? Rituxan came from Idec. Tysabri came from Elan. Tecfidera came from Fumapharm. Daclizumab came from PDL Biopharma/Abbvie. Ocrelizumab came from Genentech/Roche. Long acting Factors XIII and XI came from Syntonix Pharmaceuticals. STX-100 is a BIIB moelcule but had to leave for 5 years in order to be successfully developed by Stromedix. In the meantime the Immunology Department has produced no drugs since it’s inception in the mid-80s, well over 20 years ago. The oncology experiment (BIIB San Diego) produced no drugs. The BIIB hemophilia group will produce no new drugs (more on this below). The medicinal chemistry effort has produced no drugs (although we think they will). The BIIB neurology research group has produced no drugs outside of the interferon space, although they are getting closer (anti-Lingo, BIIB037). So why is this company even competitive, indeed dominant, in MS?

The answer is simple and compelling. BIIB excels in the development of in-licensed, clinical stage MS programs. Look at what they’ve brought in and then brought to registration: Tysabri is the single best MS drug available (nothing else is even close); Tecfidera is the single best oral MS drug, and again it’s not even close; Daclizumab will present an extraordinary efficacy/safety profile, and so on. Let’s also consider that while BIIB was accumulating and developing these assets, their competition was developing cladribine, alemtuzumab (campath), lemtrada, aubagio and other hideous potions. Even Novartis came razor close to missing with Gilenya, a nicely efficacious drug that has a challenging toxicity history

Perhaps anti-Lingo antibody will join the BIIB parade of success in MS, but company management is not counting on it. When management set a goal of “400 in 5” in 2012, they meant it, which means they cannot wait for anti-Lingo or any other early Phase 2 program to mature. This is our final point from above, that management will de-risk the pipeline. This means they have 2 choices, and they have been excellent at executing on either or both of these choices:

1) Buy a late clinical stage MS asset/company.

2) Cut costs in order to manage EPS aggressively.

A third possible outcome of course is that they will do both. A very interesting question is: what attractive MS asset/company could BIIB buy? There are some very compelling answers, and maybe we’ll share these, but not today. A less interesting question, because the answer is so obvious, is where to cut. Let’s go back to those hemophilia drugs, brought in on a wave of enthusiasm for the much broader hematology space. What happened? When costs needed to be trimmed a “strategic review” quickly revealed that hematology was not so attractive after all. So the hemophilia R&D group was slashed, and only the clinical programs retained. Note further that those Factor XIII and Factor XI drugs are utlilizing very valuable and expensive bio-manufacturing capacity for the company. What might happen here? BIIB could sell the programs for 10-20x annual sales to Bayer or Novo Nordisk and keep the manufacturing rights for 5 years or more. We’re just guessing, but we also think it’s a very good bet.

The other obvious target is the Immunology group. A possible hint here is that a new department has been formed, carrying the name Remodeling and Repair or something similar. The department is built around the very interesting Phase 2 fibrosis program STX-100, mentioned above. A simple decision would be to move the few Immunology clinical assets (the anti-TWEAK and anti-CD40L antibodies) under this new department, and jettison the Immunology Research efforts. Such a move would mimic what was done in the hematology space, and would further move the company further away from basic Research, which historically has failed to move therapeutics forward, and further toward Development: in-licensing, clinical execution, regulatory execution and bio-manufacturing, the company’s true core competencies.

Will BIIB do any of these things? We have no idea. But we have watched this company for a long time, and if top-line results fail to drive EPS to the goals promised, the company will act decisively to control the bottom line. Personally, we expect to see an acquisition in short order, rather than further cuts. Just to reinforce what we said at the beginning: the proposed corporate strategy fundamentally de-risks the impact on the anti-Lingo program on the company fortunes, leaving intact the potential for a large upside if that program performs well in the clinic.

disclosures: PDR was a senior member of BIIB’s Immunology department for a long time, and retains both positive and negative biases. PDR is also long BIIB stock.

stay tuned

Merck’s MK-3475 Deals: Assets, Risk and Innovation in Immunotherapy Pipelines

The recent news that Merck will aggressively partner the anti-PD-1 program MK-3475 with competitors Pfizer and Amgen, and biotech Incyte, was a welcome recognition that the immunotherapy landscape is too vast and complex for most companies to handle alone. Companies that will succeed in this space over the long haul will position themselves to “sample” many assets and technologies, particularly in combination settings. Why? First, because many combination therapies will fail or be too toxic to use, second, therapeutic modalities will evolve rapidly or be replaced, and third, personalized oncology practice will fragment patient populations.

Merck’s anti-PD-1 antibody MK-3475 is an example of the first generation of immune checkpoint inhibitors, for which we have clinical data. Other first generation therapeutics are ipilimumab (Vervoytm) approved for the treatment of metastatic melanoma, and nivolumab, an anti-PD-1 antibody moving toward regulatory submission this year, both from Bristol Myers Squibb. There are other anti-CTLA4 and anti-PD-1 pathway antibodies in clinical development, just a bit behind, including antibodies to PD-L1 from Roche (RG7446) and Astra Zeneca (MEDI-4736) and to CTLA4 from Pfizer (tremelimumab). It is fair to say that Merck has generated intense buzz around the MK-3475 program, driven by excellent clinical data and an aggressive approval strategy.

If we look over the details of the Merck collaborations we see a convergence of technologies around combination therapy. The Merck/Amgen collaboration centers on developing the oncolytic vaccine T-Vec in combination with MK-3475. The therapeutic hypothesis is relatively straightforward. The immune response to vaccines built using tumor antigens is blunted, in part, because of the immunosuppressive signals induced by PD-L1 expression on tumor cells. So blocking PD-L1/PD-1 mediated immunosuppression may allow a more robust immune response to anti-tumor vaccination strategies. I’ll note here that Amgen recently reported Phase 3 results from their T-Vec trial in metastatic melanoma, hitting the primary clinical endpoint of durable response but just missing the secondary endpoint of improving patient overall survival, which is an endpoint that the immune checkpoint inhibitors do hit. So Amgen has clear motivation here to combine T-Vec with immune checkpoint inhibitors. In addition to the collaboration with Merck, Amgen also has a collaboration with Bristol-Myers Squibb to clinical evaluate the combination of ipilimumab and T-Vec in metastatic melanoma.

The collaboration between Merck and Incyte is also focused on disabling immunosuppressive signaling, in this case as mediated by inhibition of indoleamine 2,3-dioxygenase (IDO), a pathway that regulates T cell responses by depleting tryptophan from the local tumor environment. IDO also appears to regulate T cell activity in lymph nodes draining the tumor site. IDO inhibitors promote T cell effector function while reducing the immunosuppressive activity of T regulatory cells. Incyte’s IDO inhibitor INCB24360 is in Phase 2 clinical trials in metastatic melanoma and in ovarian cancer. In this case then we are considering the potential of dual immune checkpoint inhibition, blocking PD-1 and IDO simultaneously. Incyte already has a Phase 1/2 trial in metastatic melanoma of INCB24360 in combination with ipilimumab and a Phase 1 trial in late stage melanoma in combination with a tumor vaccine.

Merck’s MK-3474 collaboration with Pfizer is very interesting. A phase 1/2 combination trila with axitinib a VEGFR-selective multi-kinase inhibitor (Inlytatm), will be run in renal cell carcinoma. Axitinib is approved as second line therapy in kidney cancer, but the drug has limited potential as a long term therapy and has struggled to distinguish itself from the older multi-kinase inhibitor sorafenib (Nexavartm, from Bayer/Onyx). It is very hard to guess what such a trial will yield, but such combinations of targeted therapies (kinase inhibition in this case) and immune-checkpoint modulators will have to be tried. A recent example, combining ipilimumab and the BRAF inhibitor vemurafenib (Zelboraftm, from Roche) in metastatic melanoma induced unacceptable liver toxicity and was stopped after only four patients had received the combination. Ipilimumab and nivolumab have very different toxicity profiles, and attempts at different combinations are certainly warranted.

The collaboration between Merck and Pfizer also includes development of the combination of MK-3475 with Pfizer’s PF-05082566, an agonist anti-4-1BB antibody. 4-1BB is a potent immune stimulatory pathway that acts by boosting T cell activity. Of interest, PF-05082566 is already in a Phase 1 solid tumor (as monotherapy) and B cell lymphoma (as dual therapy, with Roche’s anti-CD20 mAb rituximab). Finally, Merck and Pfizer have an second agreement to investigate the combination of MK-3475 with palbociclib, a CDK4/6 inhibitor that recently showed encouraging data in advanced breast cancer, although without yet demonstrating an impact on overall survival. These types of combinations are designed to give that precise boost in efficacy, allowing at least some patients the benefit of long term responses that impact disease progression and survival.

Merck’s internal immunotherapy pipeline is thin but as we noted the other day they are beginning to target other pathways, in part via the Agenus/4-Antibody platform deal.

I titled this post “Risks, Assets and Innovation in Immunotherapy Pipelines” because Merck’s efforts around MK-3475 illustrate some clear themes in this space. One, already mentioned, is that going into this space solo is something no company, except perhaps Bristol Myers Squibb (BMY), can contemplate. Even BMY reached outside the company recently to license an anti-KIR antibody from Innate Pharma and to partner with Five Prime Therapeutics on antibody discovery. Why is BMY in such a dominant position? They were innovative (CTLA4 biology) and they have multiple assets including antibodies to CTLA4, PD-1, LAG-3, KIR, 4-1BB and PD-L1, with more on the way. Note that I’m not saying that BMY’s anti-PD-1 antibody nivolumab is better or worse than Merck’s MK-3475, nor do I much care which gets approval first, a race that lots of folks are watching. No one horse will win this field, which brings us back to assets and pipelines.

Beyond BMY, companies like Merck are aggressively partnering because as the immunotherapy field was developing they were less innovative, took fewer risks, and therefore have fewer assets in this space. Companies like Pfizer and Novartis that spent the last decade chasing one oncogenic mutation after another down the rabbit hole found themselves very quickly on the outside looking in. They are now buying and partnering to build portfolios.

And that’s just fine; further, this should be a “pull” environment that motivates the biotech community to generate an abundance of assets. Small biotechs are classically trying to be innovative (to differentiate), take risks (to return dollars on investment) and therefore develop new assets. These are the fundamentals that should drive further expansion of the immunotherapy portfolio across the industry. So how is biotech doing in this new landscape? It is a mixed picture. There is a dearth of new first and second-generation immunotherapeutics, a space that I believe should be asset-rich. This is why Five Prime, 4-Antibody and Costim were all able to do healthy deals relatively early in their development – there is just not a lot of competition.

What happened? Why aren’t there half a dozen anti-PD-1 and anti-PD-L1 antibodies looking for partners, or a dozen agonist antibodies to 4-1BB, OX40, and GITR, and multiple inhibitors of TIM-3 and Lag3? I think this is a case of history repeating itself. After remicade and etanercept were approved, biotechs ran from the TNF inhibitor space, all believing, incorrectly, that they would never be able to compete. Seven TNF inhibitors later, this class still dominates the rheumatoid arthritis market. I wonder if small biotechs are reluctant to follow-on with additional antibodies to the first and second-generation immune checkpoint space because they think they are “too late” – in other words, the value proposition is too risky. If so, I believe they are wrong. The appetite is clearly there, with large biopharma and antibody engineering companies hungry for assets to pull into their pipelines.

Instead, many small biotechs are trying to stay well ahead on the innovation curve, chasing new targets. The problem, as always, is that no one wants to fund that work, because the risk is very high. So the answer is to try to balance innovation and risk in order to create assets that investors will fund. Its a tricky proposition, but essential to biotech’s ability to continue contributing to immunotherapy, and driving value creation. That said, there are some terrific innovative programs out there, in the hands of focused and smart small companies. In the meantime, there are more companies seeking validated assets than there are good programs developing these assets. SugarCone Biotech spends a lot of time building strategic programs for biotechs and a lot of time matching quality assets with partners and investors, so we see this first- hand.

Why else would we need more assets? Straight off, the existing immune checkpoint antibodies ipilimumab and nivolumab induce some terrific responses, but the response rates could be improved. Second, development of the existing combination therapy of ipilimumab and nivolumab has been tricky, with excellent efficacy but troubling toxicity. Note here that BMY can tinker with the dosage and dose schedule of each agent in such combination trials, because they own them both. Less asset-rich companies seeking to develop combination therapy strategies either have to partner their programs (Merck, as discussed here, but AbbVie might be another good example), or acquire everything they can afford (Novartis).

We’ll be watching closely.

stay tuned.


 by Paul D Rennert, February 11, 2014

In looking at Acute Myeloid Leukemia (AML) we see a cancer field right on the cusp of change in clinical practice. Standard of care chemotherapy regimens and stem cell transplantation protocols have proven to be of limited utility, especially in older patients. However, potentially big advances in care are being made, with exciting news coming out regularly. As we move toward the spring Medical Conference season, we felt an overview of this rapidly evolving area of oncology would be timely.

AML is a rapidly growing cancer of myeloid lineage cells that proliferate in the bone marrow and interfere with normal hematopoiesis. AML typically arises in the context of defined genetic mutations. For example, translocations of chromosome 16 disrupt RUNX1 gene activity and are one of the several underlying causes of Core Binding Factor AML. CBF-AML). Since RUNX1 regulates the transcription of many genes, the effect of its disruption is complex. CBF-AML patients are generally responsive to chemotherapy initially, although up to half of these patients will relapse over time due to additional genetic mutations.

Mutation of the FLT3 protein is the most common genetic abnormality in AML, found in about 30% of patients. This is a genetic characteristic associated with poor prognosis. The most common FLT3 mutation, FLT3-ILD, is caused by an tandem duplication within the coding region of the gene. The resulting protein drives hyper-signaling and oncogenic cell responses. Mutations that change the active site of the protein, causing unregulated phosphorylation, have also been described. Mutations in the receptor tyrosine kinase c-Kit are also associated with oncogenic signaling in AML. Both of these pathways cause mutiple downstream effector pathways to be activated. The JAK2 mutations, commonly see in myelofibrosis and other myeloproliferative disorders, are rare in AML but when characterized can potentially be treated with Jak2 inhibitors.

According to a recent market research analysis                           ( a total of 62,226 new cases of Acute Myeloid Leukemia (AML) were recorded in 2010, with 95,000 predicted new cases for 2015 and nearly 130,000 predicted new cases in 2020. Note that as of February 2014 approved agents for AML remain limited to chemotherapeutics ( Despite the lack of new targeted drugs, the AML therapeutics market was nearly 240 MM USD in 2011. At the current rate of growth the AML market could reach over 700 MM USD by 2018. These numbers are based on the analysis of future AML drugs growing at a 17% compound annual growth rate from through 2018.

Numbers like these are continuing to drive intensive research into effective, novel therapies for AML. It only helps that in many cases such therapeutics find use in other hematopoietic diseases such as Chronic Myeloid Leukemia (CML) and in the B cell lymphomas, including Hodgkin’s Lymphoma and the non-Hodgkin’s Lymphomas (NHL).

 There is obvious unmet medical need for effective therapies in AML since this is a disease characterized by quick relapse after therapy with grim survival statistics. In some older patients, survival is as little as 1-1.5 years despite first and second line treatment regimens.

What’s exciting from the drug development and biotech investment perspectives is that the AML treatment landscape is advancing simultaneously across therapeutic modalities. This rapidly changing landscape give us a chance to look at targeted small molecule drugs, monoclonal antibodies (naked, bi-specific, radiolabelled, immunotherapeutic, ADC), targeted T cells and other novel technologies.

 We can then ask ourselves: who will the winners be in 5 years?

 A) Targeted small molecule drugs.

Lets just be clear upfront that the goal of these targeted therapies is to get patients who have relapsed, or are refractory to chemotherapy, to a complete response (CR) with minimal residual disease (MRD) so they can qualify for an allogeneic stem cell transplant (SCT). That’s a lot of acronyms but what this is really saying is that for most patients the goal is a modest one – we are not asking for a durable remission, at least not yet.

 A variety of established drugs are being tested in AML. Also, the identification of oncogenic mutations in FLT3 and cKIT has driven interest in developing new tyrosine kinases inhibitors (TKIs) for AML.

 Sorafenib (NexavarTM; Bayer and Onyx) is a dual targeting drug that blocks RAF signaling (and therefore the MEK>ERK signaling) and also the growth factor receptor tyrosine kinases VEGFR and PDGFR. The NCI is running a large phase 3 trial enrolling new onset pediatric AML patients (NCT01371981) with sorafenib being given in combination with various chemo regimens.  Bayer and Onyx are running several earlier phase AML trials. An interesting phase 1 trial in patients 18 or older combines sorafenib with plerixafor and G-CSF (NCT00943943). The idea here is to have the CXCR4 blocker (plerixafor) and the growth factor (G-CSF) flush tumor cells, and also tumor stem cells, from the bone marrow and lymph nodes so that they are more sensitive to sorafenib treatment. This trial is co-sponsored by Genzyme/Sanofi, which owns plerixafor.

Another interesting trial is the Phase 1/2 study of the combination of sorafenib, with vorinostat, and bortezomib (NCT01534260). Here we have a proteasome inhibitor and an HDAC inhibitor added to growth factor and signaling inhibition provided by sorafenib. This potent combination is being used in patients with a poor genetic risk profile, including FLT3-ILD positive tumors. This study is co-sponsored by Bayer/Onyx, Millennium/Takeda and Merck Sharp & Dohme Corp.

Bristol Myers Squibb is running an interesting trial (NCT01620216) in which AML and acute lymphocytic leukemia (ALL) patient samples are analyzed for sensitivity to drug treatment ex vivo, after a period on drug in the trial, as follows:

“An in vitro kinase inhibitor assay will be used to determine the sensitivity of primary leukemic cells to four kinase inhibitors/drugs:

Drug: Sunitinib, 50 milligrams (mg) qd, with or without food, for 4 weeks

Drug: Dasatinib, 100 mg qd…possible escalation to 140 mg qd for 28 days

Drug: Nilotinib, 400 mg twice daily for 28 days

Drug: Sorafenib, 400 mg (2 tablets) orally twice daily without food for 28 days

Drug: Ponatinib, 45 mg dose once per day

Sunitinib (Sutenttm; Pfizer) makes sense as a pan-growth factor receptor inhibitor; dasatinib (Spryceltm; Bristol Myers Squibb) is a Src and c-Kit inhibitor and is a reasonable choice for AML; nilotinib (Tasignatm; Novartis) is a pretty specific Bcr-Abl kinase inhibitor and is probably only being used for the ALL population – and even there only 25% of ALL patients carry this translocation; sorafenib we discussed earlier; ponatinib (Iclusigtm; Ariad) has a grab bag reactivity profile, hitting the BCL-ABL kinase, FLT3, RET, c-KIT and the FGFR, PDGFR and VEGFR growth factor receptor kinases. This is a dangerous drug, with a very narrow FDA approval in CML, and I suspect enrollment in this little exploratory trial will be stopped if possible.

If I had to guess I would say that this rather odd trial design has several goals. One is to look for signs of efficacy, although a month is pretty short duration. One might also look for patterns of resistance to therapy, which would be very interesting. Since this is BMY, I’d be surprised if they weren’t also looking at cell surface markers for possible immunotherapy treatment – more on this subject later.

Results from a dasatinib trial in CBF-AML were recently presented at the American Society of Hematology (ASH) conference (Abstract #357). Dasatinib was added to induction and consolidation chemotherapy in newly diagnosed AML patients. Unlike the rrAML population, the CBF-AML population can experience sustained periods of remission prior to relapsing, especially in younger patients. Since some of the relapses are driven by gain of function mutations in c-Kit, dasatinib should prevent at least those clones from becoming established. Early results looked good but longer term data are needed to see if this regimen will remain effective.

Imatinib (Gleevectm; Novartis) another Bcr-Abl, c-Kit and PDGF-R inhibitor, has been tested in multiple AML trials, but the results have not led to approval for use in AML. An interesting trial of the cytotoxic/immunomodulatory agent lenolidomide (Revlimidtm; Celgene) plus chemotherapy is being run by the NCI (NCT01246622). Lenolidomide has been approved for the treatment of a different bone marrow resident cancer, multiple myeloma (MM).

Anyway there is a lot of similar clinical trial work being done – using approved drugs in this new indication and looking for efficacy. This is ultimately good both for patients and the drug development companies.

Lets move on to some newer drugs in the pipeline. The FLT3 inhibitors give us a sense of the difficulty here, with low response rates as monotherapies.

Quizartinib (Ambit Biosciences; AMBI) remains stuck between phase 2 and 3 for relapsed/refractory (rr) AML. This drug is a FLT3 inhibitor with a somewhat tortured history, having been partnered for a time with Astellas, then returned, then running nicely in the clinic before running into disagreement with the FDA over approvable endpoints and safe dosage. In early December the company announced it would have to run a phase 3, likely with lower starting doses, in order to obtain FDA approval. Investors were hoping the company could file on its phase 2 trials. Notably, later in December Ambit showcased its’ quizartinib data from the FLT3-ILD rrAML trial, in which a 50% response rate (50% or greater reduction in leukemic blast cells) was reported with relatively low doses of drug. Unfortunately, it will be a while yet before more news becomes available about this drug.

In the meantime heavy hitter Novartis is already in phase 3 with its’ FLT3 and Protein Kinase C inhibitor midostaurin. The phase 3 in newly diagnosed patients is being run by the NCI (our tax dollars at work), along with The Alliance for Clinical Trials in Oncology and the Cancer and Leukemia Group (NCT00651261). A trial of midostaurin administered with or without bortezomib in adult rrAML patients is being run by Novartis and Millennium/Takeda (NCT01174888). Preliminary results were presented at ASH (abstract #3966). While response rates were impressive the toxicity was extreme, and this seemed to be due to the bortezomib dose, which was adjusted. Phase 2 trials in adult patients who carry c-KIt, FLT3-ILD, and various other mutation or cytogenetic markers are also underway (NCT01830361, NCT01846624). A phase 2b midostaurin  monotherapy study published several years ago showed modest improvement in AML patients with mutated FLT3; this study recognized the need for combination therapy to improve the clinical response (

Another FLT3 inhibitor, lestaurtinib, is the subject of 2 NCI sponsored trials in pediatric ALL/AML but drug development of this agent seems to have stalled when Teva bought Cephalon. Another FLT3 inhibitor is PLX3397 (Plexxikon) which has activity against  KIT, CSF1R and FLT3. This drug is in a phase 1/2 trial in adult rrAML (NCT01349049).

One of the major challenges for FLT3 inhibitors is breadth of action. These inhibitors work best on patients who have mutated FLT3 and are less effective in patients with normal FLT3. Also, secondary mutations have already been discovered in response to FLT3 inhibition. Specifically, in those patients who have mutations in the active site of the kinase, so-called gatekeeper mutations arise, conferring resistance to the drug.

A dominant theme in recent drug development for AML has been built on the observation that proteasome inhibitors can impact cancers of the bone marrow. Disruption of proteasome activity blocks a wide spectrum of cellular activities, and is particularly effective against rapidly dividing cells (like leukemic blasts) but also relatively quiescent tumor stem cells, that require specific proteasome-dependent signaling pathways (e.g. NK-kB). Bortezomib (Velcadetm, Millennium/Takeda) has shown activity in older patients when combined with chemotherapy. A phase 3 combination trial with sorafenib in newly diagnosed AML patients is underway, sponsored by the NCI (NCT01371981).

Carfilzomib (Onyx Pharmaceuticals) is in an early stage trial for AML, along with extensive trials in MM, B cell lymphomas, etc. The drug is furthest along in MM, now in phase 3 (NCT01568866). Early reports so far have suggested that this drug has an activity profile similar to bortezomib, but may have a better safety profile. This is an interesting drug (and company) to watch. They have a second generation oral version of carfilizomib, oprozomib, in phase 1 MM trials. Millennium/Takeda are developing ixazomib in MM and lymphomas. An AML trial is listed but not yet recruiting.

A third theme that we can follow in AML therapeutic drug development is the use of drugs that impact epigenetic gene regulation. Because AML is driven by genetic translocations, gene regulation at the level of chromatin structure is disrupted. There are two processes at work here that can be targeted. One is the aberrant methylation of CpG islands in gene promoter regions, which can be targeted by DNA methyltransferase inhibitors. The second is changes in the conformation of chromatin caused by dysregulated histone acetylation. This process can be therapeutically targeted using histone deacetylase [HDAC] inhibitors.

The HDAC inhibitor vorinostat (Zolinzatm, Merck) has been extensively studied in AML, and is currently in a phase 3 trial with chemotherapy for young patients with newly diagnosed disease (NCI; NCT01802333). Vorinostat monotherapy was generally ineffective, but combination with chemo agents proved much more potent. As detailed at ASH in December (Abstract #2684), newly diagnosed and rrAML patients were enrolled in a phase 2 expansion study. Of 75 patients, 57 patients achieved CR, and 7 achieved CR with incomplete platelet recovery (CRp), for an overall response rate of 85 percent. Median overall survival was 82 weeks and median event free survival was 47 weeks. For patients with the high-risk Flt-3 ITD mutation the 10/11 achieved CR and 1/11 CRp. The ORR = 100% in these patients. Their median overall survival was 91 weeks and median event free survival was 66 weeks. About 25% of the total patients in CR received SCT.

Other combination trials include the sorafenib trial mentioned above, and a trial in combination with antibody therapy (gemtuzumab ozogamicin) for rrAML (NCI; NCT00895934). This trial reported early results at ASH (Abstract #3936). The response rates ere encouraging and about 20% of patients obtained durable remission. There were significant toxicity issues. This drug is very likely to play a critical role in the evolution of combination therapy for AML. We’ll discuss antibody therapies further in Part 2.

Other important HDAC inhibitors in development for AML is panobinostat (Novartis). What’s interesting about the development campaign with this drug is the pairing in multiple trials with 5-azacitidine, a DNA methytransferase inhibitor. In such settings two modes of epigenetic regulation are being targeted simultaneously. One of these studies published findings last month                                 ( and demonstrated good tolerability and reasonable response rates. Clearly, this combination should move forward in the context of chemotherapy or other drugs. Of note the DNA methyltransferase inhibitor decitabine (Dacogentm, MGI Pharma) is already approved for AML. There was also a presentation on the HDAC inhibitor entinostat (Syndax Inc) with 5-azacitidine in myeloid neoplasia (Abstract #2777), and there are several clinical trials listed for AML, however this drug is mainly being used in solid tumor trials.

Other interesting drugs in this area include alisertib, an Aurora A kinase inhibitor (Millennium/Takeda) being tested extensively in B and T cell lymphomas and in solid tumors. There are several AML trials including a phase 2 trial completed by MLMN (NCT00830518). Selinexor, (Karyopharm) a selective inhibitor of nuclear export, in in phase 1 trial for advanced AML. Abbvie’s Bcl2 inhibitor ABT-199 is also in an AML trial.

If we take a step back we can appreciate that in small molecule development Novartis, Merck and Onyx are placing big bets in this therapeutic area. We’ll sort out the best looking therapeutics as we dig in a little deeper.

In Part 2 we’ll take a look at the biologics landscape, and begin to draw the bigger picture.

The development of combination therapies for B cell lymphoma: ABT-199

The role of ABT-199 in the development of combination therapies in lymphoma.

Following yesterday’s blockbuster win for PCYC and JNJ – the Phase 3 trial versus Arezza was stopped early on clear PFS and OS benefit – it seems a little deflating to return to AbbVie, whose Bcl2 inhibitor ABT-199 has been dogged by Tumor Lysis Syndrome (TLS) problems, some fatal, and recent rumors of oversight problems at one clinical site. I stated the other day that ibrutinib would win the medical marketplace for B cell lymphoma treatment, based on its impressive suite of clinical trials, and the results announced yesterday support that opinion. I still believe that combo therapy is the critical path forward in this field, and ibrutinib and idelalisib are the clear leading candidates for combo treatment protocols.

However, ABT-199 remains a wildcard and could be transformative if developed carefully. The drug demonstrates ORR and CR responses in B cell lymphoma that are very dramatic, as detailed earlier. In the spirit of our earlier January posts, lets look at the clinical trial spectrum for ABT-199. There are five trials listed for ABT-199 monotherapy in oncology, including the phase 1 extension. These include trials in relapsed/resistant NHL, CLL, high-risk CLL (del17p), MM, and AML. The inclusion of acute myeloid leukemia (AML) distinguishes ABT-199 from the other lymphoma drugs, and is based on the mechanism of action and profiling of different tumor types for sensitivity to Bcl2 inhibition.

The combination trials are very narrow in scope and reflect the fact that the ABT-199 is partnered with Roche/Genentech, and therefore the anti-CD20 mAbs used are those developed by Roche, that is, rituximab and obinutuzumab (Gazyva). Obinutuzumab is already approved for the treatment of previously untreated CLL, i.e., as first line therapy. The obinutuzumab trial with ABT-199 is sponsored by Roche/Genentech, illustrating the depth of this collaborative effort. Roche has an ongoing preclinical BTK program, and it will be interesting to see if a combination trial with ABT-199 is eventually filed.

Here are the combination trials listed:

Trial NumberPhasedate filedCombinationIndication
NCT0159422914/20/2012bendamustine/rituximabrrNHL and DLBCL
NCT0167190418/10/2012bendamustine/rituximabCLL (rr & untreated)
NCT016826161b6/26/2012rituximabrrCLL and SLL
NCT0168589219/12/2012ObinutuzumabCLL (rr & untreated)

The ultimate success of the program will depend in large measure on controlling the TLS issue. Its worth reminding ourselves that TLS did not suddenly become a toxicity concern in the treatment of B cell lymphoma. It occurs with anti-CD20 treatment and with chemotherapy treatment, as a result of triggering the death of a large number of tumor cells. The problem for ABBV and ABT-199 is that TLS does not seem to be related to dose given, remaining somewhat unpredictable (aside from worrying when patients present with bulky disease, which indicates huge number of tumor cells in the lymph nodes, bone marrow, etc). The recent outcry over moving one patient from one dose to another (150mg to 1200mg if I remember right) seems a little silly when a dose of 50mg can trigger TLS in a lymphoma patient. That said, the burden is on AbbVie to demonstrate that they can provide better efficacy than the competing drugs (ibrutinib and idelalisib), safely. If they can do this, I predict that ABT-199 will have a big role to play in the treatment of B cell lymphoma. If they continue to struggle then this drug will will still have a role, but may be relegated to second line or even salvage therapy status. Given the resources behind it, from both AbbVie and Roche, I imagine that a huge effort will be brought to bear on understanding and controlling the TLS toxicity.

Final ASH13 SnapShot: AbbVie’s ABT-199

Update on ABT-199. 12-3-2013, by @PDRennert 

I reviewed the status of ABT-199 back in August, following ASCO (see link). At the time I felt the drug was being overlooked in the hype around ibrutinib and idelalisib. This was based on impressive response rates and the sense that AbbVie had gotten a handle on how to dose safely. After all, tumor lysis syndrome (TLS) is not uncommon in the treatment of lymphomas, and can be dealt with by dose modification or intervention.

At ASH we get a further look at this drug. Abstract #872, to be presented by J. Seymour, introduces a modified dosing regimen. Patients (rrCLL/SLL, n = 56) were given single doses of ABT-199 at day -3 or day -7, then started on daily dosing after that. This is a Phase 1 monotherapy trial designed to determine MTD. The efficacy readouts were pretty dramatic. ORR = 84% with CR = 21% and PR = 63%. Within the CR group (n = 12), 8 patients had no or very low minimal residual disease. There were 12 discontinuations due to progressive disease (21%). The response rate was not related to del(17p) status but the PR rate was lower in fludarabine-refractory patients (these patients are all at high risk).

Notably, among the usual AEs (diarrhea, neutropenia, URIs, etc) there was an 11% grade 3/4 TLS rate. This is 6 patients. Problematically, TLS was not clearly dose associated: 2 @ 50mg, 1 @ 100mg, 2 @ 200mg and 1 @ 1200mg, this last one resulting in sudden death. So dose optimization remains in progress.

The observation that response was not related to del(17q) status or other aberrations in the TP53 locus in rrCLL patients will be discussed in more detail by M.A. Anderson (Abstract #1304).

M. Davids (Abstract #1789) will present a similar study in a variety of rrNHL patients, including MZL, MCL, DLBCL, FL and WG. Of the 32 patients enrolled and followed for a median time of 6 months, 18 discontinued due to progressive disease (that’s 56%). AEs were typical (nausea, diarrhea, cytopenias, URIs). There were 2 mild episodes of TLS. For FL and DLBCL, doses of 600mg or higher were required for efficacy. The responses were good given this difficult mix of patients. Note that the text and table in this abstract don’t entirely line up, so best to hear the current results at the meeting. The implication here however is that ABT-199 is safer in these patient populations than in rrCLL, a theme we also heard at ASCO.

Aside from the clinical data there is an awful lot of preclinical modeling using cell lines or patient derived cells in vitro or in vivo (mouse). The point of all this nice work is to show the potential of combination therapy using ABT-199 along with other drugs. A few examples:

–  2-DG + ABT-199 kills all Myeloma subtypes (#1921)

–  ibrutinib + ABT-199 is effective against MCL cells and CLL cells (#645 and # 3080)

–  imatinib + ABT-199 kills chronic phase CML cells

–  BTK inhibitor R406 + ABT-199 kills DLBCL cells

   etc, etc

The CML observation points to another trend, which is efficacy of ABT-199 in settings beyond NHL, including ALL (#3919), AML (#885) and MM (#4453). There are others…

On balance this remains an exciting and potentially important drug. The issue of TLS in certain subtypes of NHL remains to be solved, while in other, difficult NHLs there appears to be clear and compelling efficacy with less toxicity.

Apoptosis Induction for the Treatment of B Cell Lymphomas: Update on AbbVie’s Bcl-2 Inhibitor ABT-199.

Lost in the fanfare surrounding the BTK inhibitor Ibrutinib and the PI3Kdelta inhibitor Idelalisib at this year’s American Society of Clinical Oncology meeting (ASCO 2013) was some rather stunning data from AbbVie on their Bcl-2 selective inhibitor ABT-199.
ABT-199 binds to Bcl-2 in such a manner as to prevent this protein from interacting with pro-apoptotic proteins such as BAX and BID, thereby allowing them to attach to the mitochondrial membrane, induce cytochrome-c release and trigger apoptotic cell death. Tumors disable the Bcl-2 pathway in a number of clever ways. Many tumor types delete p53, a protein that normally up-regulates expression of BAX and BID and down-regulates expression of Bcl-2 and related proteins (BclxL, Mcl-1, Bcl-w, etc). Tumors with this deletion phenotype are termed 17p(del) referring to the site of chromosomal alteration. Such tumors are generally very aggressive, resistant to chemotherapy and radiation, and associated with poor prognosis.
B cell lymphomas constitute a diverse collection of lymphomas and some leukemias (the distinctions have become blurred). B cell lymphomas have historically been classified as Hodgkin lymphomas or the very diverse Non Hodgkin lymphomas (NHL). The term NHL is confusing as this covers chronic lymphocytic lymphoma (CLL), mantle cell lymphoma (MCL), diffuse large B cell lymphomas of 2 subtypes (DLBCL-ABC and DLBCL-GCB), indolent Non Hodgkin lymphoma (iNHL) and a host of other tumor types. A baffling mix of biology and acronyms, but simply put, all of these tumor types are derived from various stages of B cell development, and most require signaling through the B cell receptor (BCR) to survive. The BCR triggers proliferative and survival (anti-apoptotic) signals, and therefore Bcl-2 becomes a relevant target in NHL.
Not surprisingly, genetic evaluation of B cell lymphomas has revealed mutations in Bcl-2 or downstream of Bcl-2 that foster lymphoma cell survival. Examples include constitutive activation of Bcl-2 in follicular lymphoma (FL) and CLL, amplification of Bcl-2 in MCL, down-regulation of NOXA (another pro-apoptotic protein) and BIM (a protein that regulates Bcl-2 activity) in iNHL and CLL, and down-regulation of expression of the caspases, part of the apoptotic machinery triggered by mitochondrial cytochrome-c release, in CLL.
ABT-199 was developed as a follow-on to the earlier Abbott compound navitoclax (ABT-263). Navitoclax inhibited multiple Bcl-2 family members, including BclxL, and was associated with severe thrombocytopenia (platelet loss) in clinical trials. Bcl-xL is an obligate pro-survival protein for platelets and thrombocytopenia became a dose limiting toxicity in navitoclax clinical trials. ABT-199 has a much lower affinity for BclxL than for Bcl-2, and spares other proteins in this family, as shown in this data reproduced from a recent AbbVie paper (Souers et al. 2013. Nat Med 19: 202-210):
                                 TR-FRET Ki (nM)
                      Bcl-2    Bcl-xL    Mcl-1    Bcl-w
navitoclax       0.044    0.055      > 224     > 7
ABT-199      < 0.01     48          > 444     245
What is important to note here is the relative difference in potency of ABT-199 against Bcl-2 (less than 10 picomolar, an unbelievable potency number) vs 48nM against BclxL. Taken at face value this is a 4800 fold difference in target potency. As noted below however, the number may be deceiving given the drug exposure achieved in patients.
AbbVie presented several studies at ASCO, including their Phase 1b trial in CLL. What is striking about the data is the response rates, as these compare favorably with similar data shown recently for BTK inhibitors and PI3K inhibitors, from Gilead (Idelalisib), Infinity (IPI-145), Celgene (CC-292) and J&J (ibrutinib). A snapshot of results from the CLL trials is shown below.







PI3Kd isoform
PI3Kd/g isoforms
Partial response (PR) 

40%      (n = 5)
Complete response (CR) 

Overall response (OR = PR + CR) 

60% (n=5)
OR in 17p(del)
The table is adapted from results presented at ASH 2012, ASCO 2013 and this year’s European Hematology Association meeting (EHA 2013), as generously posted on Twitter by @andybiotech. ABT-199 differentiates from the competing drugs in several critical ways. First the OR rate reached 84% and was not statistically different between CLL patients having or not having the p53 deletion, ie.17p(del). Second, the CR rate was 18%, meaning that this many patients absolutely cleared disease from the bloodstream, lymphatic fluid, and lymph nodes. In addition AbbVie stated at ASCO that in the CR group, 11% had complete recovery of the bone marrow and 7% had partial bone marrow recovery – these are responses that are not seen with BTK or PI3Kg/d or PI3Kd inhibitors.
So why did the fanfare around NHL treatment coming out of Chicago in May during ASCO 2013 not include ABT-199? The answer was toxicity, and this toxicity came in 2 distinct forms. One has been addressed recently, the other is somewhat complex.
Tumor Lysis Syndrome (TLS) is a drastic physiological response to the sudden and acute destruction of massive numbers of tumor cells, more or less simultaneously. The disgorging of massive quantities of intracellular material triggers acute physiologic response as the heart, kidney, spleen and other organs cope with a titanic overload of potassium, calcium, phosphate, and uric acid. Organs fail and circulation collapses the patient drops quickly toward death. ABT-199 caused this syndrome in the first 3 patients that were dosed, each having been given 200mg of drug. One died. Faced with this spectacular disaster AbbVie immediately halted all ABT-199 clinical trials. That was back in January or February of this year. So the update at the end of May was tempered by this ongoing toxicity issue, and what AbbVie had to say at that point didn’t help. Starting with a lower dose of 50 mg they had restarted all trials, and gotten the drug into 34 CLL patients. Three experienced TLS, 1 died, 1 lost renal function (acute renal failure, generally meaning a lifetime of dialysis or a transplant). In addition, other toxicities were present, as seen in the next table showing severe toxicities (grade 3+) in CLL clinical trials. URI: upper respiratory infection. NA: data not available







PI3Kd isoform
PI3Kd/g isoforms

2% (URI)
18% (pneumonia
19% (pneumonia)
Liver damage (AST/ALT) 




The table is adapted from results presented at ASH 2012, ASCO 2013 and EHA 2013, as generously posted on Twitter by @andybiotech.
Where really jumps out here is at the level of thrombocytopenia and neutropenia seen in patients receiving ABT-199. Its worth noting that patients taking any of these new drugs are already at risk for decreased cellularity due to chemotherapy, and so adding to this burden complicates their subsequent care. Neutropenia is a bona fide pathway toxicity as demonstrated by defective expression of Bcl-2 in Kostmann syndrome (severe congenital neutropenia) and so there is nothing to do about this but lower drug exposure, potentially at the expense of efficacy. An interesting question in this regard is whether the thrombocytopenia and anemia seen are due to residual inhibition of BclxL. If you look at the PK data presented from the CLL trial, its pretty clear the effective concentrations of drug are far in excess of what is needed to inhibit BclxL. Taking data presented by Seymour et al. at ASCO 2013, abstract #7018, its clear that the maximum concentration achieved after multiple doses (Cmax(ss)) is up to 2mg/ml, which is going to give low mM exposure. So, at Cmax and for some time thereafter, drug concentration exceeds that required to inhibit BclxL by as much as 50-fold. Whether the duration of inhibition is sufficient to induce thrombocytopenia and anemia in patients is not known.
Lets go back to TLS, as this is the real show-stopper. On their July 26, 2013 earnings call, AbbVie stated that the dosing regimen has been further modified (details not clear). AbbVie further stated “With regard to tumor lysis syndrome which is a direct consequence of the explicit potency of 199 we have been enrolling CLL patients with a revised dosing schedule where we start at a lower dose and ramp up at a more slow rate, and so far so good with regard to the patients that we have been treating under that new protocol.” It is beginning to look as if AbbVie is learning how to manage TLS, and therefore have the potential to get their drug back in the fast lane.
Why does all this matter? Selective Bcl-2 inhibition is a unique strategy and ABT-199 illustrates compelling activity. In the era of rapidly advancing combination therapy for NHL, this could be an important component of the evolving treatment paradigm. With a lower dose they will try to avoid TLS and perhaps blunt thrombocytopenia and neutropenia. In the near term this drug may play a prominent role in patients who are at high risk (17p(del) or who are refractory to other targeted therapies. More interestingly, Bcl-2 inhibition in combination therapy may be a breakthrough treatment paradigm, and in this regard, AbbVie is swinging for the fences with trials combining ABT-199 with R-CHOP (Rituxan + chemotherapy) in pursuit of achieving outright cures. Combine this with genetic profiling of Bcl-2 and related proteins across many tumor types, and you have a very interesting story. 

This is a drug to watch, and just, perhaps, to improve upon.
As always, stay tuned, and follow us on Twitter @PDRennert.