The FDA’s decision to reject Roche’s (RHHBY.PK) filing for accelerated approval of T-DM1 had quite an impact on Immunogen’s (IMGN) stock. There is no doubt that T-DM1 represents the company’s most valuable program, even if it is in the form of a mid single digit royalty rate, due to its blockbuster potential and impressive clinical activity. This justifies to some extent last week’s market reaction, as the next potential approval for T-DM1 is anticipated in mid 2012. Nevertheless, T-DM1 is facing multiple potential value creation events in the coming year, including an important data set for T-DM1 next month. In addition, the company is involved in 6 clinical stage programs (#7 is expected to enter the clinic this month), some of which are expected to generate data in the coming months. Although none of these programs is nearly as exciting as T-DM1, some of them could become more attractive with time.
T-DM1 related events
It is important to note that the regulatory setback has nothing to do with T-DM1’s activity or Immunogen’s technology. T-DM1 is still a promising drug with impressive activity and a very good safety profile in heavily pretreated breast cancer patients. The next shot is in 2nd line approval where Roche is conducting a large double arm trial head to head vs. standard of care. In this case, the trial is probably well powered to support filing providing the data is positive. Naturally, large randomized controlled studies are always more challenging compared to single arm trials, but givenT-DM1’s clinical profile, the drug has a decent shot to beat the control arm (Tykerb+Xeloda).
T-DM1’s potential market remains unchanged, with a blockbuster potential based on the $4.8B in sales Herceptin had last year. Herceptin’s sales will probably grow beyond $5B following the approval in HER2 positive gastric cancer. As I explained two years ago, T-DM1’s potential could be even larger than that currently served by Herceptin as there are several additional indications where a subset of patients express HER2. In these indications, Herceptin was not successful but since T-DM1 has a totally different mode of action, it could work where Herceptin failed, just like in Herceptin pretreated breast cancer.
The vast majority of Immunogen’s value is derived from T-DM1, which is in a broad and aggressive development program. The pivotal data in 2nd line patients will probably be available late next year. Until then, T-DM1 will have readouts from several early stage trials.
The most important of these readouts will occur next month at the ESMO meeting, where investigators will present a trial with T-DM1 in 1st line patients (not previously treated with Herceptin). This will not only be the first opportunity to evaluate T-DM1 in Herceptin naïve patients but also the first trial that compares T-DM1 with another treatment regimen (Herceptin+Taxotere). It will be important to see how T-DM1 performs in comparison to an established chemotherapy regimen, and based on Roche’s decision to initiate a phase III trial for T-DM1 in 1st line patients, there should be at least a comparable level of activity.
Other interesting trials involve combining T-DM1 with other agents. Preliminary results from one of the trials evaluating T-DM1 and pertuzumab were reported earlier this year. Although the safety profile of the combination was good, the efficacy seemed similar to that of single agent T-DM1 in prior trials. Interestingly, the 1st line pivotal trial includes a T-DM1+pertuzumab arm, so perhaps updated results from the phase I will be better. Another interesting study combines T-DM1 with GDC-0941, a PI3K inhibitor. This combination is interesting because preliminary signs show that over activation of the PI3K pathway renders tumors less sensitive to T-DM1 and other HER2 targeting drugs.
The rest of Immunogen’s pipeline
There are currently 6 agents in clinical development that are based on Immunogen’s technology. Based on the available data, none of them comes close to T-DM1 in terms of clinical data and likelihood of approval. Of course, new data (good or bad) will change the inherent value of these compounds.
IMGN901 (anti CD56, wholly owned)
Results from 2 trials are expected this year. The first trial is an expansion of an ongoing phase I study in a rare type of cancer called Merkel Cell. This indication was chosen based on unusual anecdotal activity in Merkel Cell that was discussed here. As I wrote, this could be a very fast route to market and a meaningful value creation event but results to date are ambiguous. Updated data is expected at ESMO next month and will enable the company to decide whether or not to pursue development of IMGN901 in this niche indication.
The second trial is a phase I in multiple myeloma that will probably be reported at ASH in December. I reviewed earlier results from this trial here. The updated results will give a better sense of the agent’s clinical profile but the data is unlikely to justify pursuing IMGN901 as monotherapy in multiple myeloma. Immunogen has already initiated a combination trial with Revlimid, which is a common drug for this indication. Preliminary data from this trial will not be meaningful as it will be hard to associate activity with IMGN901 directly.
SAR3419 (anti CD19, Sanofi-Aventis (SNY))
Last year Sanofi presented positive (but not overwhelming) results from a phase I of SAR3419 in lymphoma. At ASH this December, the company will present data from another phase I utilizing a more frequent dosing schedule (weekly administration). It appears that this regimen is more active than the prior regimen and Sanofi plans to use the weekly regimen in future phase II studies. This, however, resulted in a delay in the initiation of SAR3419’s phase II, which is now expected to begin in the summer of 2011. This is good news for Micromet (MITI), who is developing a competing antibody for the same target, as I discussed last year. Seattle Genetics (SGEN) is also expected to enter the clinic with its anti-CD19 ADC next year.
IMGN388 (anti alphav integrin, wholly owned)
This ADC was originally developed by J&J’s (JNJ) Centocor, who decided not to advance it into clinic trials. Immunogen saw this as an opportunity to bring a compound to the clinic very quickly and with minimal cost. There are several antibodies that target integrins in development, as well as two approved antibodies in non oncology indications. Although integrins have an established role in cancer associated functions such as invasion and motility, the value of anti-integrin antibodies in cancer is still unclear.
Alphav integrin is found in many solid tumors and tumor blood vessels, but it is also present on normal cells. This target is being pursued by Pfizer (PFE) and Centocor through naked antibodies. The Centocor antibody (CNTO-95) has been in clinical trials for 8 years and is currently in a phase I study where it is given together with Avastin. Results to date were unimpressive although there were some signs of clinical activity as a single agent.
IMGN388 is interesting because it is the first and only antibody drug conjugate targeting an integrin. The naked antibodies aim at disrupting the biologic activity and signaling of integrins and so far showed little promise. As an ADC, IMGN388 uses integrins for delivering a drug payload into cancer cells. On top of the activity issue, there were doubts about the safety profile of IMGN388 because integrins are expressed by certain normal cells as well.
3 months ago, Immunogen released first data from a phase I trial that showed the ADC is surprisingly safe. There were no responses, even though some of the patients’ tumors had high integrin expression. There were several cases of disease stabilization as well as what could be seen as a correlation of integrin expression and dose to clinical benefit, but numbers were too small to draw definitive conclusions. Immunogen will release an update from the trial in November. Expectations from this agent are low, so any clinical activity seen with IMGN388 will be a positive surprise.
BT062 (anti CD138, Biotest)
Biotest is developing BT062 for multiple myeloma. Results to date indicate certain degree of activity but similarly to IMGN901, this agent is not likely to be pursued as a single agent. Last month, Biotest announced its plans to start a trial evaluating more frequent administration.
BIIB015 (anti-CRIPTO, Biogen Idec (BIIB))
This ADC entered phase I more than two years ago, however, no results were published, which is usually not a good sign. Expectations from this agent are low.
SAR650984 (Anti CD38, Sanofi Aventis)
This is a naked antibody for blood cancers which recently entered clinical testing. Genmab is developing its own CD38 antibody in multiple myeloma but has not reported anything from a study that is ongoing for almost 3 years. Morphosys has a preclinical program for CD38 as well.
SAR566658 (anti DS6, Sanofi Aventis)
SAR566658, which will probably enter clinical testing in a matter of weeks, is an ADC targeting DS6. DS6 a sugar side chain on MUC1, a well known cancer associated target. To date, all attempts to target MUC1 failed, and ADCs are not considered a good option for this target due to lack of specificity. SAR566658 targets a side chain on MUC1 that seems to be more differentially expressed. Immunogen already used this approach with IMGN242 which is an ADC against a MUC1 glycoform (CanAg), but its development was discontinued last year due to insufficient efficacy. In any case, meaningful data for SAR566658 is not expected in the near term future.
Business development activity
Immunogen is expected to announce new licensing agreements after a long drought of 2 years with no new deals (the last deal with Bayer announced in October 2008). This is surprising given the strong validation T-DM1 has provided to Immunogen’s technology. On the recent conference call, CEO Dan Junius addressed this discrepancy, explaining the company decided to be more selective in its deals due to the need to retain enough resources for internal development. In addition, it seems Immunogen is asking for terms that are substantially better than those of its current collaborations.
In contrast to Immunogen, Seattle Genetics has seen quite a lot of business activity, including three broad collaborations with GSK, Genentech (DNA) and Agensys (part of Astellas (ALPMF.PK)) as well as a licensing deal for lead agent SGN-35 with Millenium (Takeda (TKPHF.PK)). The four deals brought Seattle Genetics over $100M in upfront payments.
Similarly to Immunogen, Seattle Genetics derives the vast majority of its valuation from a single drug, SGN-35, which is expected to be approved next year for Hodgkin’s Lymphoma. Results from a pivotal phase II trial are expected next month. This is also a single arm phase II but unlike the T-DM1 trial, it is conducted under a SPA (special protocol assessment), which minimizes the risk of last minute surprises from the FDA.
Seattle Genetics’ pipeline is expanding at a fast rate, including phase I initiations for 7 different ADCs in the past year or so. One of these agents is wholly owned by the company (SGN-75), two are being co-developed with Agensys(AGS-5ME, AGS- 16M8F) and four are being developed by partners. It seems that two of these programs have already been terminated: MEDI-547’s trial has been terminated prematurely probably due to toxicity. Another ADC that was developed by Genentech entered phase I in 2008 but there was no mentioning of this agent ever since.
In summary, both Immunogen and Seattle Genetics have one promising agent in development and an expanding pipeline of early stage ADCs to which little value can be ascribed at the moment. It is also likely that the majority of these agents will fail to reach the market, based on the high attrition rate in oncology. The good news is that positive data from any of these early stage programs will have a big impact on the company’s valuation. While it’s impossible to predict which programs will become successful, the growing number of ADCs in the clinic make this a matter of “when” rather than “if”.
Disclosure: Long IMGN, SGEN, MITI