Below is a recap of interesting data expected to be presented at the Annual meeting of The American Society of Clinical Oncology (ASCO). This year’s meeting will have a strong focus on immunotherapy (PD-1 antibodies in particular) as well as novel oral agents for hematology. Companies for which important data are expected are reviewed as well.
PD-1 takes center stage
PD-1 antibodies are by far the hottest and most active field in cancer drug development. Although technically, all data to date were from phase I studies, there is wall to wall consensus that inhibitors of the PD-1 pathway will become a cornerstone in several cancer types and a $5-$10B franchise.
The undisputed leader is BMS’ (BMY) nivolumab, which is in 6 pivotal trials. The two closest competitors are Merck (MRK) and Roche, who are in phase II. The BMS and Merck antibodies bind PD-1 whereas Roche’s antibody binds PD-1 ligand (PD-L1). All three agents already generated exciting data and the main question is whether these agents are interchangeable or not (i.e do they have the same clinical profile?).
Data at ASCO could start to answer this question although it is unlikely to see unequivocal proof until late stage results are available. Given BMS’ clear early mover advantage, Roche and Merck are trying to differentiate their programs as early as possible. Focus is expected to be on cross-trial comparison per indication, activity spectrum in new indications, safety profile and biomarkers for patient selection.
PD-1 Data highlights
Combination of nivolumab and Yervoy (anti- CTLA-4, also from BMS) in melanoma – Response rate of 47%-50% at relevant doses appears numerically higher than that documented with either drugs alone but the sample size is small. BMS will start phase III trial for the combination later this year, which indicates updated results will be positive as well. BMS is the only one with an approved CTLA-4 antibody on the market, which could help it to corner the market, at least in melanoma.
Update from nivolumab’s phase I trial in melanoma, NSCLC and RCC – Although technically designated phase I, this was a huge study of over 250 patients. Response rates continue to look strong. As with Yervoy, there appear to be patients who achieve long term remissions that could eventually be seen as cure.
Median survival for melanoma, NSCLC and RCC respectively were 16.8, 9.6 and >22 (not reached) months respectively. These values appear substantially better than historical data, although it is hard to interpret survival data from single arm trials. Initial results in PD-L1-positive melanoma patients also look promising with a dramatic difference in overall survival between PD-L1 positive (21.1 months) and PD-L1-negative (12.5 months).
Update from MPDL3280A’s phase I – The abstract discloses good activity (21% response rate) across various tumor types. Activity was seen in the 3 “usual suspects” (melanoma, NSCLC, RCC), generally in line with nivolumab’s early data but surprisingly, activity was seen in tumor types where nivolumab was not effective (colon cancer) or has not been evaluated (gastric, head and neck). This potentially unique activity spectrum could be an important differentiator for Roche, however, results are still very preliminary.
Another interesting point is that BMS also had a PD-L1 antibody (MDX-1105), which demonstrated lower response rates than nivolumab and MPDL3280A. The basis for this discrepancy is unclear although it appears that patients in the MPDL3280A trial were less heavily-pretreated compared to BMS’ trials. There is obviously the possibility that MPDL3280A is simply a better antibody.
One potential advantage of targeting PD-L1 is lower toxicity. MPDL3280A does not inhibit PD-L2 binding to PD-1, which may be associated with some of the pneumonitis (lung inflammation) seen with PD-1 antibodies (fatal is some cases). Indeed, to date no severe pneumonitis was reported with MPDL3280A, although follow up is still limited.
Abstracts also show clear correlation between PD-L1 expression and response across tumor types (39% of PD-L1–positive had a response vs. 13% for PD-L1–negative patients). This was more pronounced in the NSCLC cohort, where all 4 patients with PD-L1-positive tumors responded compared to 15% of PD-L1–negative patients. Needless to say, numbers are still small.
PD-L1-positive NSCLC will probably be Roche’s first indication based a single arm phase II trial expected to open this month. >25% response rate in this molecularly defined subpopulation should be enough for accelerated approval. BMS is also pursuing accelerated approval in NSCLC, but they focus on squamous histology rather than PD-L1 expression.
Update from Merck’s lambrolizumab (MK-3475) in melanoma– Data available only in melanoma patients, where response rate (>35%) is similar to slightly better than other PD-1 inhibitors. The abstract can be viewed as disappointing, as a previous update reported a response rate of 51%. This might be due to the fact that the prior update included also unconfirmed responses.
Merck is pursuing lambrolizumab in a large randomized phase II trial in pre-trated melanoma patients (Including Yervoy failures).
Seattle Genetics – More ADC data from Roche
Seattle Genetics’ (SGEN) investors should track 2 data sets from Roche’s (Genentech) antibody-drug conjugate pipeline. As a reminder, Roche has 8 ADCs in clinical development, all of which utilize Seattle Genetics’ technology. As more data for these programs becomes available, the market is starting to realize the tremendous value these 8 programs represent. Today it is clear Seattle Genetics is much more than Adcetris but an ADC powerhouse with substantial economic interest in multiple partnered programs.
Until now, Roche unveiled positive phase I results for 3 ADCs targeting CD22, CD79b and MUC16 (Discussed here). In all 3 cases, a good safety profile and robust efficacy signals were noted. At ASCO, 2 additional ADCs will have initial results that appear positive based on the abstracts. This brings the number of partnered programs to which investors can ascribe value to 5.
DNIB0600A (anti- NaPi2b) is in phase I for NSCLC and ovarian cancer. The abstract discloses 3 confirmed responses (still ongoing) and 1 unconfirmed response among 18 patients who received higher doses.
DSTP3086S (anti-STEAP1) is in phase I for prostate cancer. The abstract discloses a relatively modest activity profile but there were encouraging signs of CTC (circulating tumor cells) reductions that seem dose dependent.
Seattle Genetics’ agreements with Genentech entail a total of $1.4B in milestones and mid -single digit royalties per program. If only one or two of Genentech’s programs reach late stage testing, it should have a material impact on Seattle Genetics’ valuation (Best example is Immunogen’s mid single digit royalty stake in Kadcyla).
Additional partnered ADC programs to be presented at ASCO are Progenics’ (PGNX) anti-PSMA for prostate cancer and Genmab’s anti-TF (preclinical data).
Immunogen – First look at IMGN853
Immunogen (IMGN) will present phase I results for IMGN853 (anti-folate receptor), which is emerging as the company’s most important internal program. The company is going into ASCO with a positive sentiment towards IMGN853, which still needs to be corroborated by actual data.
There are high expectations around IMGN853 based on positive results from Endocyte’s (ECYT) Vintafolide (also targets folate receptor) in ovarian cancer. Vintafolide is a small-molecule drug conjugate and has inferior pharmacologic properties compared to ADCs, so IMGN853 may be more potent. Vintafolide is no active as a single agent, so if IMGN853 generates responses, it should be viewed as a superior competitor.
At ASCO, investors will look for preliminary signs of activity in the form of tumor shrinkage and CA125 reduction. The abstract has very limited data on efficacy but 1 case of an 82% reduction in CA125 in a patient receiving a high dose (3.3 mg/kg) is mentioned. The fact the drug could be escalated to 5 mg/kg is encouraging as these are levels where activity is typically observed with ADCs.
Exelixis – Promising activity in RET mutated tumors
Exelixis (EXEL) will have several clinical updates for cabozantinib (Cometriq) in different indications. Although none are expected to have a major impact on the stock, they could provide some read-across for the ongoing phase III in prostate cancer as well as insight on activity in RET-mutated tumors.
Survival data from prostate cancer phase II – Updated results from a single-arm phase II in prostate cancer will be reported. By definition, analyzing results from non-comparative trials is problematic but the trial was fairly large (144 patients). As background, this was the trial that demonstrated the drug’s controversial bone scan normalization effect (reviewed here).
The abstract reports a median survival of 10.8 months, which is encouraging given the tough patient population (73% were ≥3rd line, many were taxane resistant/refractory). The most important read across is for the ongoing phase III trial, as investors will try to put cabozantinib’s survival in context with Zytiga’s and Xtandi’s post-chemo survival (with the caveats of cross trial comparisons).
Since patients in cabozantinib’s phase III are Zytiga and/or Xtandi failures, cabozantinib’s phase II data can be compared to the post-treatment survival in the phase III studies for both agents (8.5-10 months). At ASCO, survival for the Zytiga pre-treated subgroup will be particularly relevant as this subgroup more closely resembles the population being enrolled in the Phase 3 trial.
At first glance, cabo’s 10.8 months does not look impressive but it appears cabozantinib phase II had a much tougher patient population based on prior therapies and resistance to taxanes (Taxotere or Jevtana). In particular, all patients had progression within 6 months of last taxane dose and more than a third of patients progressed within 1 month after last taxane dose (highly aggressive tumors).
RET-mutated NSCLC as a new niche indication – Newly identified RET mutation (1-2% of NSCLC) represents a new niche indication for cabozantinib, which is also a RET inhibitor. At ASCO, a group from Memorial Sloan Kettering will report updated clinical experience with cabozantinib in this rare subset. Initial findings from this trial were recently published and showed promising activity in 3 patients with RET mutation, who appear to derive benefit from cabo (2 partial responses, 1 durable SD, see figure from the article below).
Source : Drilon A. Cancer Discov. 2013 May 7. [Epub ahead of print]
RET-mutated NSCLC represents a new opportunity for cabozantinib, currently not in anyone’s models. Assuming updated results are positive, Exelixis could pursue accelerated approval based on a small single-arm phase II trial. Since the drug is already approved for medullary thyroid cancer (MTC), physicians might use it off label in lung cancer. Challenges related to this program are the need for a validated diagnostic test for patient selection (Required by the FDA) and potential competition from other RET inhibitors, especially Ariad’s (ARIA) Iclusig which is also a highly potent RET inhibitor.
Subset analysis in RET mutated MTC – Subset analysis from cabo’s phase III trial in MTC shows dramatically better activity in RET mutated patients compared to non-mutated patients (PFS of 60 vs. 25 weeks). The drug is approved in MTC regardless of RET status, but this type of analysis could make physicians more inclined to use it in mutated patients. It could become even more important once overall survival data matures since the prior analysis found no benefit in the entire patient population. If the strong PFS signal in RET mutated patients is translated to a survival difference, it should bode well for Exelixis’ marketing efforts even if this does not go into the label.
Infinity –competition intensifies
Infinity (INFI) lost half of its market cap in less than 2 months (it is still over $1B), which is largely attributed to concerns about competition in CLL and NHL. As background, Infinity’s IPI-145 belongs to a new class of oral agents that are about to revolutionize the way many blood cancers are treated. The 3 most advanced are ibritunib (Btk inhibitor, Pharmacyclics/ J&J), idelalisib (PI3K-delta inhibitor, Gilead) and ABT-199 (Bcl-2 inhibitor, Roche/Abbvie). IPI-145, a PI3K gamma/delta inhibitor, is fourth in line, slightly behind ABT-199 and 1-2 years behind ibrutinib.
ASCO will have updates for IPI-145 (oral presentation for NHL) and competing drugs in CLL and lymphoma. While Infinity’s abstracts did not include new data, the competing molecules had positive updates which curbed investors’ enthusiasm regarding IPI-145.
Phacyclics’ (PCYC) ibrutinib – Continues to be the leader with “jaw dropping” results in CLL, multiple breakthrough designations and a very clean safety profile. Activity in CLL include objective responses in 67-81% of a very long progression free survival across different subsets and treatment lines. IPI-145 has a different, perhaps complimentary mode of action, but some claim it will be hard to gain market share or even recruit CLL patients for clinical trials once ibrutinib is on the market next year. A potential differentiator for IPI-145 could be activity in patients who progress on ibrutinib, which may be presented at ASCO.
Gilead’s (GILD) idelalisib – This is Infinity’s most direct competitor (target the same pathway) which is further along in development (multiple phase IIIs ongoing). Idelalisib is clearly an active drug but it is not as potent as ibrutinib and also appear to have some safety issues (potential liver toxicity). Until recently, IPI-145 appeared to be more effective, at least in CLL (55% response rate vs. 24% for idelalisib) and safer. However, updated results in idelalisib’s abstract show that response rate improved with time and it now stands at 56% (using a less strict criteria). This was the main trigger for the diminishing excitement around IPI-145, which derives most of it value as a better next-gen idelasilib.
Roche’s/Abbvie’s (ABBV) ABT-199 – This drug started to gain the market’s attention only recently, and initially was not perceived as a threat to IPI-145 due to unclear efficacy and a problematic safety profile (deaths due to tumor lysis syndrome [TLS]). 2 ASCO abstracts show robust efficacy in CLL (85% response rate) and NHL (55%) and a surprisingly good safety profile. It also appears that investigators were able to find a regimen that minimizes TLS. Phase III trials are expected to begin shortly.
Where does it leave Infinity?– Although the market is more competitive than expected, IPI-145 is still promising and Infinity could find multiple differentiation routes. Comparison with idelalisib should be done after ASCO, where results will include higher doses and longer follow up, both expected to improve activity over time. IPI-145 may also be relevant in indications that are irrelevant to its competitors such as T cell and Hodgkin’s lymphoma. Additional trials in inflammation are ongoing (results expected next year), to which the market doesn’t ascribe real value. Lastly, assuming available preliminary results are corroborated in later trials, IPI-145 has a very mild safety profile, which could be an advantage over idelalisib and ABT-199 (but not ibrutinib which looks extremely safe).
Synta – Updated survival data in NSCLC
Synta (SNTA) will present updated results from the GALAXY1 trial, a large randomized phase II evaluating its Hsp90 inhibitor (ganetespib) in NSCLC. Results have important implications for an ongoing phase III trial that was designed based on initial phase II data.
Preliminary results from GALAXY1 (which I discussed here) demonstrated a survival trend, especially in patients with a 6-month follow up. While the previous analysis included only 77 patients, the upcoming analysis should include all or most of the 250 adeno patients enrolled in the trial.
The past month provided some mixed signals about the data. Earlier this month Synta announced the departure of Dr. Sumant Ramachandra, President of R&D for “personal reasons”. This raised concerns in the market as some viewed it as a red flag, several weeks before ASCO. On the other hand, the presentation was chosen by ASCO as a Clinical Review (CRA) presentation, which means results will be released only on the day of presentation. This is typically viewed as a positive indication although it is unclear how much data the organizing committee had when it made its decision (Synta disclosed that an OS analysis is expected sometime prior to ASCO). Another positive sign was the initiation of the phase III trial, implying the survival trend was not completely lost.
At ASCO, investors will evaluate the OS curves for the entire patient population as well as patients with indolent disease (6 month from diagnosis). The latter subgroup was chosen by Synta as the phase III population due to a robust efficacy signal. Given the nervousness around Synta, any attenuation of the survival signal could have animpact on the stock.
Additional presentations include results from 2 investigator-sponsored trials for ganetespib in esophagogastric and prostate cancer, respectively. Both studies had disappointing results but there was a single case of a dramatic complete response which is still ongoing after 27.5 months. Interestingly, the patient harbored a KRAS mutation, a known client protein of Hsp90. KRAS mutations are rare in gastric cancer (0-10%), and this case demonstrates again that Hsp90 inhibition might be very effective in niche indications (such as ALK and RAF mutated NSCLC).
Roche – Protecting the CD20 franchise
Roche will report phase III results for obinutuzumab (GA101), an anti-CD20 antibody viewed as Rituxan’s successor. Rituxan (also an anti-CD20 antibody) is one of Roche’s top selling drugs (over $6.5B in sales) and several companies intend to launch biosimilar versions of the drug. Ideally, Roche should replace Rituxan with GA101 in as many settings as possible in order to remain the dominant CD20 player in hematology.
At ASCO, investigators will present phase III results from 2 separate trials evaluating GA101 or Rituxan, respectively, when added to chemotherapy in CLL. Both agents (GA101 and Rituxan) led to superior PFS over chemotherapy alone with GA101 demonstrating a numerically superior PFS (23 vs. 15.7 months). Although this result is not a direct proof for GA101’s superiority over Rituxan, it is a good indication for that.
Roche will also have results from 2 ADCs (discussed above for Seattle Genetics) and ABT-199 (co developed with Abbvie, discussed above as competition for Infinity))
Additional companies to watch
Array – Astrazeneca’s (AZN) selumetinib, a MEK inhibitor licensed from Array (ARRY), will have phase II data in uveal melanoma. There are no details as results are embargoed and will be released only at the day of presentation. This implies results are positive to some extent and may result in another phase III for selumetinib.
Array’s second MEK inhibitor, MEK162, (developed by Novartis [NVS]) will have results from a combination trial with Novartis’ BRAF inhibitor. The two drugs appear combinable at their single agent doses and preliminary signs of activity appear encouraging.
Curis – Phase I results for the IAP inhibitor, CUDC-427 (GDC-0917), will be presented. Curis (CRIS), which licensed this drug from Genentech earlier this year, is facing a lot of skepticism due to Genentech’s decision to externalize the program. As promised by Curis’ management, the drug showed initial activity signs including 2 complete responses (ovarian cancer and MALT lymphoma) of 42 patients. Safety profile remains an issue due to a fair amount of toxicity, which could be an issue especially for combination regimens.
Genmab (GEN.CO) – The abstract for daratumumab (anti-CD38) does not reveal a lot of new data from the phase I in multiple myeloma. The antibody, which was recently granted breakthrough designation, led to 5 PRs and 3 minor responses in 12 patients at the higher doses. Authors rightfully concluded: “This is unprecedented for single-agent mAb treatment of MM.”
Incyte –Investigators evaluated Jakafi’s effect on bone marrow fibrosis in patients who participated in the drug’s phase I/II trial. After 2 years, improvement or stabilization of fibrosis was more common in patients treated with Jakafi compared to patients who were treated with hydroxyurea. This is a small and retrospective trial, but bodes well for Incyte’s (INCY) effort to convince the market Jakafi is not just a symptomatic drug but a potential disease modifier.
Arqule – Arqule (ARQL) will present phase II results for its MET inhibitor ,tivantinib, in colon cancer. Although the trial did not lead to a meaningful PFS improvement in the overall population, an encouraging trend in survival emerged (HR = 0.67). More important is subset analysis based on MET status. If positive, it will be the 3rd time tivantinib demonstrates better results in MET high tumors. The company already disclosed that retrospective analysis from the failed lung cancer study also generated a survival difference in MET-high patients. This should be seen as a positive read-through for the ongoing phase III in liver cancer.
We are adding a new position in Infinity in anticipation of strong results at ASCO and a second position in ArQule based on enhanced confidence in the liver cancer trial and early data with ARQ092 (Discussed here).
Portfolio holdings – May 26th 2013