BMS’ PD-1 antibody – As good as it gets
The biggest news at this year’s ASCO came from BMS’ (BMY) PD-1 antibody, BMS-936558. This antibody belongs to a new class of antibodies that stimulate patients’ immune system to attack cancer. This approach has been recently validated with another BMS antibody, Yervoy, which was approved last year for melanoma.
Based on results presented at the meeting, BMS-936558 is superior to Yervoy by any measure. In fact, it is probably one of the most promising oncology drugs ever to be tested in humans. It induces tumor shrinkage in a substantial portion of patients, creates an immune response that keeps the disease under control for long periods and it does so with limited side effects. To make things even better, there might be a way to pre-select patients who are more likely to respond to this agent.
Data was presented in 3 indications: lung cancer, melanoma and renal cancer. In all three cases, BMS-936558 had remarkable activity in heavily pre-treated patients with objective responses in 18%, 28% and 27%, respectively. Response rates in these settings are usually below 10%, even for highly active chemotherapy drugs.
As I discussed in my pre-ASCO post, the real benefit with this agent is probably greater than demonstrated by responses, as immunotherapy drugs tend to build a prolonged effect even in patients who did not achieve response. In addition, the analysis included patients who had received suboptimal doses of the drug.
At the moment, there are 3 main antibody franchises in oncology: CD20 antibodies (blood cancers), HER2 antibodies (breast and gastric cancer) and VEGF antibodies (primarily lung and colon cancer). Each of these segments is generating ~$6B in sales every year. PD-1 antibodies will probably be the 4th on that list with sales potential of ~5-7B in just the three relevant indications.
Immunogen – Strong results for T-DM1
Roche’s T-DM1, powered by Immunogen’s (IMGN) antibody-drug conjugate (ADC) technology was the star of the plenary session, with spectacular efficacy in breast cancer. As expected, T-DM1 led a statistically significant improvement in progression-free survival (PFS) vs. Xeloda+Tykerb in 2nd line HER2+ patients (9.6 vs. 6.4 months). The positive surprise was an interim survival analysis which showed an impressive trend in favor of T-DM1. Although the data set was premature to officially declare victory, there is little doubt that T-DM1 leads to a survival benefit. This is exceptional given that unlike most cases, it was not evaluated vs. placebo but an approved regimen of 2 drugs.
The T-DM1 results serve as another validation for Immunogen’s ADC technology, as this is the first time an ADC shows a survival benefit in phase III. Besides T-DM1, the company has 10 programs in its proprietary and partnered pipelines, but to date most have not generated strong efficacy data. SAR3419, an anti-CD19 ADC developed by Sanofi (SNY) generated positive results in lymphoma patients but this agent is not as impressive as T-DM1.
Looking at ADC companies such as Immunogen and Seattle Genetics (SGEN), it is now clear that finding good targets for ADCs is more challenging than originally thought. However, T-DM1 serves as proof that once the right target is found, the clinical effect is dramatic. It is therefore reasonable to assume that the “next T-DM1” is only a matter of time.
Many ADCs in Immunogen’s pipeline are still early in development and the market ascribes minimal value to them. Next year, this could change as the company and its partners are expected to report clinical data from several programs. If one ADC demonstrates similar activity to that of T-DM1in its phase I, the impact on the company’s valuation could be substantial.
Interestingly, it appears that one of Immunogen’s program, a naked CD38 antibody in development by Sanofi, has the potential to be a value driver for the company in 2013. Although no data has been reported to date, Genmab released impressive results with its own CD38 antibody in multiple myeloma (to be discussed below).
Array Biopharma – Maturing into a phase III company
Array (ARRY) had a transformative meeting thanks to positive results from its 2 MEK inhibitors, selumetinib and MEK162. Both drugs will probably advance to phase III trials in the coming 12 months.
Selumetinib, in development by AstraZeneca (AZN), had an impressive effect in lung cancer with KRAS mutations when added to standard chemotherapy. The drug increased response rate dramatically (35% vs. 0%) and PFS was more than doubled (5.3 vs. 2.1 months), compared to chemo alone. Both differences were statistically significant.
Overall survival also increased dramatically (9.4 vs. 5.2 months) but the difference was not statistically significant, which is not surprising given the trial size. Based on this remarkable data set, which is in a prospectively predefined patient population, a phase III looks imminent.
MEK162 also generated a promising signal in a molecularly defined subset – melanoma with NRAS mutations. These patients (~15% of total patients) represent a highly unmet need as they are not candidates for BRAF inhibitors such as Roche’s Zelboraf. Of 28 evaluable patients, 18 had some level of tumor shrinkage, including 6 partial responses (3 confirmed). Although this activity is not as impressive as the ~50% response rate seen with BRAF inhibitors in BRAF mutated melanoma, it still merits pursuing MEK inhibitors in this subset of patients who have very limited treatment options. (Alone or in combination with other drugs).
Array is looking at an active remainder of 2012. As the field for MEK inhibitors is crowded (the leader is GSK’s trematinib), both AstraZeneca and Novartis will probably outline their registration strategies for their respective drugs later in 2012. The company also expects to present updated results for its 2 hematology drugs at ASH in December.
Another catalyst this year could come from ARRY-797, a p38 inhibitor for pain, currently in a randomized phase II in osteoarthritis of the knee. Although no data have been published from the trial, p38 has been recently validated as a target for osteoarthritis of the knee by Flexion.
Flexion announced that its p38 inhibitor was efficacious in a randomized phase II study in osteoarthritis of the knee. Flexion’s approach differs from that of Array, as its drug is injected directly to the joints whereas Array’s drug is given as a pill. Flexion’s data implies that if ARRY-797 manages to reach the joints, it has a good chance of showing efficacy. Based on prior ARRY-797 trials, it has a very good safety profile, which could be an important distinguishing factor.
Exelixis – Results continue to look strong
ASCO was a good meeting for Exelixis (EXEL) as cabozantinib (cabo) continued to generate impressive data in prostate, renal and liver cancer. Despite the compelling data across several indications, there is still a great deal of skepticism on Wall Street regarding cabo. Critics argue the bone effect might not be real or clinically relevant, and that the mechanism of the drug is not well understood. While these are valid points, it is hard to ignore the extent of cabo’s clinical activity and the unique profile it has, not observed with any other drug to date.
In prostate cancer, where cabo is currently in 2 phase III trials, investigators presented phase II results in 93 patients who had failed chemotherapy. The drug led to a bone scan response (partial or complete) in 67% of patients and appeared active in soft tissue lesions as well. The dual activity on both bone and soft tissue lesion is a unique feature of cabo, not observed with any other drug in prostate cancer. While this profile bodes well for cabo’s ability to prolong survival in phase III, it could be an important differentiator in the highly competitive prostate cancer market.
In renal cancer, cabo had remarkable efficacy in a small 25-patient phase II trial. Patients in the study were heavily pre-treated, two thirds of which had received 2 or more prior treatment lines. Half of the patients had a history of a VEGFR inhibitor (such as Sutent) and an mTOR inhibitor (Afinitor).
Cabo led to a median PFS of 14.7 months, which is very encouraging for this patient population. To put things in perspective, approved 2nd line treatments such as Afinitor or Inlyta have a PFS of 5 months. In this trial, most patients were 3rd line patients who exhausted approved treatment options. Cabo had a response rate of 28%, again substantially exceeding that of approved 2nd line drugs (2% for Afinitor).
While in prostate cancer cabozantinib is a class of its own (not in direct competition with other agents), in renal cancer it will compete with other kinase inhibitors. The RCC field is packed with VEGFR inhibitors including Pfizer’s (PFE) Sutent and Inlyta, GSK’s (GSK) Votrient and Aveo’s (AVEO) tivozinib. All these drugs are direct competitors with limited differentiation and patients treated with one VEGFR inhibitor derive limited benefit from another VEGFR inhibitor. Pressure on new drugs is only expected to increase after Sutent, the mainstay for 1st line renal cancer, goes off patent. Based on the limited data, there is a chance cabo could be potent enough to become an effective 2nd line treatment or even the leading 1ST line drug. Exelixis plans to initiate a 150-patient randomized phase II trial in 1st line patients vs. Sutent.
Arqule – Survival signal in liver cancer
Arqule (ARQL) started ASCO with a negative sentiment based on a modest PFS signal from a phase II in liver cancer. In MET+ patients, tivantinib led to a PFS benefit of almost 1 month, which is probably not considered clinically meaningful. The updated overall survival data, however, showed a dramatic separation of the curves (median survival of 9 vs. 3.8 months). The main caveat for this analysis is the very low sample size (28 patients) but given statistically significant signal in PFS and the prospective biomarker analysis, this is enough to push the drug to phase III in MET+ liver cancer.
Genmab – Potential breakthrough in myeloma
Genmab presented phase I results for daratumumab, an anti-CD38 antibody for the treatment of multiple myeloma. Daratumumab was given to very heavily pretreated myeloma patients, including patients with a history of 10 (!!!) or more previous treatment lines. Data included multiple objective responses as well as some dramatic reductions in myeloma cells in the bone marrow. Responses appear dose dependent, with 5 partial responses and 2 minor responses in the 12 patients who received the 4 highest doses (response rate of ~42%). This compares to no PRs and 2 minor responses among the 17 patients at the lower doses.
This type of activity is very rare in multiple myeloma, especially in such a challenging patient population. Importantly, daratumumab’s mechanism of action is distinct from the two classes of drugs that are active in myeloma: IMiD’s (Revlimid, pomalidomide) and proteosome inhibitors (Velcade, carfilzomib). Therefore, it is not a direct competitor of these agents and will probably be added to standard therapy.
In fact, daratumumab is the only antibody to date to show such a robust efficacy profile in myeloma. For example, BMS’ elotuzumab, currently in phase III in multiple myeloma leads to no responses as monotherapy. Even antibody drug conjugates such as Immunogen’s IMGN901 demonstrate a response rate of 5-8%.
The only issue with daratumumab was the fact that it was given with dexamethasone in order to minimize side effects. Dexamethasone is known to have activity in myeloma, but the extent of activity and the dose dependent response profile imply daratumumab has intrinsic activity.
Genmab’s results are good news for Immunogen and Morphosys, which also have anti-CD38 antibodies in clinical trials. Immunogen’s antinody is licensed to Sanofi whereas Morphosys fully owns rights to its CD38 antibody, MOR202. Sanofi is evaluating SAR650984 in several CD38+ blood cancers whereas Morphosys is focusing on multiple myeloma. CD38 antibodies could represent an important class of drugs with limited competition and a combined market opportunity of $2-3B in myeloma alone.
Seattle Genetics – Label expansion efforts
Although Seattle Genetics did not get a lot of attention this year, the data presented for Adcetris continues to support expansion to additional treatment settings. Adcetris’ was active when given to patients who had already been treated with the drug, with an impressive response rate of 70%.
The drug was also active in DLBCL, with a 57% response rate among patients who express Adcertis’ target, CD30. Assuming this level of efficacy is corroborated in larger studies, this creates an opportunity for Seattle Genetics to pursue approval based on a single arm phase II trial, duplicating its initial strategy in hodgkin’s lymphoma. Since only a portion of DLBCL patients express CD30, the company will have to use and validate its companion diagnostic to get approval.
In an effort to identify additional relevant patients, the company published preliminary results from its screening study, where patients with solid tumors are evaluated for CD30 expression. Several interesting subsets were described, but no efficacy data was reported in the meeting.
Although Synta (SNTA) did not present important clinical findings at this year’s meeting, the company generated a lot of interest among investors. The primary catalyst for its Hsp90 program, ganetespib, is preliminary results from a randomized phase II trial in lung cancer expected later this month. The company already announced the analysis identified certain patient populations for which ganetespib appears beneficial. These subtypes will be evaluated in a phase III program expected to start later this year.
Synta is evaluating the drug across multiple indications in an attempt to identify genetic profiles which are “addicted” to Hsp90 inhibition. So far, Hsp90 inhibitors generated an efficacy signal in ALK mutated lung cancer and HER2-positive breast cancer. The company presented preclinical data on 2 new profiles in lung cancer that were sensitive to ganetespib. These new profiles, defined by ROS1 and RET fusion mutations, represent two distinct niches in lung cancer.
Synta’s most advanced competitor is Novartis, which is developing AUY922, an Hsp90 inhibitor it licensed from Vernalis. At ASCO, Novartis presented results in lung and breast cancer, demonstrating clinical activity in some patients. Interestingly, AUY922’s activity in different subsets (EGFR and KRAS) was different from what has been observed so far with other Hsp90 inhibitors. AUY922 led to visual disturbances in the majority of patients but Novartis is moving ahead with a broad development program, as most cases of visual side effects were not severe.
Debiopharm and Astex (ASTX) also presented results for their respective Hsp90 inhibitors. Debiopharm’s compound (Debio0932), licensed from Curis (CRIS) demonstrated some efficacy in unselected patients (2 PRs in 45 evaluable patients). Notably, this drug can be given orally, in contrast to all the other Hsp90 inhibitors. Astex’s AT13387 also led to a fair amount of visual toxicities and generated early signs of activity.
Portfolio holdings as of June 10th, 2012