This year is shaping up as a transformational year for Synta (SNTA), which is making progress on multiple fronts with its lead agent ganetespib (formerly known as STA-9090). Ganetespib is an Hsp90 inhibitor, a protein with a well recognized role in cancer. The concept of inhibiting Hsp90 to fight cancer goes back two decades, but all attempts have been beset by failures so far. As Synta’s ganetespib appears to be the first active and safe Hsp90 inhibitor, it is poised to make a big dent in the multibillion dollar oncology market. I discussed the history of Hsp90 inhibitors and Synta’s unique positioning in a previous write up.
Synta started off the year with good data in non-small cell lung cancer (NSCLC) and expects to accelerate ganetespib’s development toward a registration trial early next year. In the meantime, ganetespib is being evaluated in a broad and aggressive clinical development program, which could generate a lot of clinical data in 2011. Lastly, Synta is expected to partner for Asian rights for the drug in the coming months. This deal should be very lucrative, as ganetespib represents everything the pharma industry is looking for: Clear signs of activity, established safety profile, broad potential utility and biomarkers for patient selection.
Promising Clinical Data
Last month, Synta published results from a phase II trial in NSCLC. The long anticipated data included results for 33 patients whose tumors did not harbor EGFR and KRAS mutations. These patients represent about two thirds of the overall population with NSCLC, the leading cause of cancer-related death in the western world. There were three cases of partial responses as well as seven additional cases of tumor shrinkage. These results are in line with the previous experience with Infinity’s (INFI) first generation Hsp90 inhibitor in NSCLC. The problem with Infinity’s drug, which is based on a natural antibiotic, was its problematic safety profile, primarily liver toxicities.
Based on these results, it is now safe to say that ganetespib is definitely an active drug as a single agent. This is further supported by responses in a variety of other indications, for which ganetespib is currently being evaluated. According to the company, it is seeing responses in breast, gastric, melanoma and colorectal cancer, some were in very heavily pretreated patients. This demonstrates the broad potential utility ganetespib possesses, which is not surprising given the central and ubiquitous role Hsp90 has in cancer biology.
Still, ganetespib did not have a meaningful effect in the majority of NSCLC patients and this is probably the case in other tumor types. This activity profile is typical of targeted therapies, which do not shrink tumors but curb their growth in the majority of patients. Some targeted agents, such as Tarceva and Nexavar, lead to a survival benefit as single agents in unselected patient populations despite their mild activity profile. Nevertheless, in order to prove this for ganetespib, Synta will have to run a long, expensive and risky study.
Two additional alternatives for developing targeted agents are identifying patients who are more likely to respond using biomarkers or combining targeted agents with approved regimens. Synta is pursuing both directions in its recently announced registration program in lung cancer.
Last month, Synta announced plans to initiate a phase 2b/3 study in NSCLC in combination with Taxotere, a commonly used chemotherapy drug. This trial, planned to start in the next quarter, will include a phase 2b portion that will generate a go/no go decision for the phase III portion and will inform the company on the optimal trial design. The phase III portion, which will involve the regimen and medical centers from the phase II portion, will start only based on good phase II data. The uniqueness in Synta’s approach is in coupling the phase II with the phase III from operational and regulatory aspects. This will save the company time and money as well as facilitate a smooth transition between the two phases. Needless to say, if the phase 2b fails, this design has no apparent advantage.
There are basically two important questions the phase II portion should address. First, it will serve as an indication for the benefit in combing ganetespib with Taxotere. Another important aspect of the phase 2b trial will be identifying biomarkers in order to better define the patient population most likely to benefit from ganetespib. This is a crucial point as it will enable Synta to pursue approval of ganetespib as a highly effective treatment for niche populations, not necessarily in combination with Taxotere. This may represent a fast and cheap route to market on top of the long and standard phase III evaluation, assuming both are viable options.
The issue of biomarkers for patient selection has been touted by the company for a while but no specifics were provided. It appears that the company has already identified genetic markers in the patients who had dramatic responses with ganetespib but this data is still preliminary. There are probably multiple mutations that render cancer cells “Hsp90 addicted." Even when assuming that these mutations occur in a small precentage (~10%) of the population, the cumulative market size across the different tumor types could be substantial. One marker that is already out in the open is ALK mutations in lung cancer patients, which I discussed in my recent post on Synta.
Synta is using PFS as a primary endpoint in the phase II portion, which is typically short in second line NSCLC (3-4 months with Taxotere alone). Consequently, initial results from the phase 2b trial are expected late 2011 or early 2012, less than a year from study initiation.
The trial is not blinded, which somewhat compromises the robustness of the data. The advantage in not being blinded to the data is the ability to monitor in real time the trends and adjust the clinical strategy accordingly. Another potential issue might stem from using PFS as a basis for decision making given the relative fluid nature of this endpoint, as opposed to overall survival. In addition, given the short PFS expected in the control arm, it might be challenging to show a meaningful numerical difference.
With the available data set, which proves ganetespib is both safe and active as a single agent, the time is right for a partnership deal. Management did not provide specific guidance but insinuated that the preferable scenario will be a deal for marketing rights in Asia at this point followed by an additional deal down the road when the program is more advanced.
Based on licensing deals in past years, ganetespib could be the subject of a massive licensing deal. The table below lists seven licensing deals of targeted agents for cancer between biotech companies and large pharmas. What all these drugs have in common is the fact they target hot pathways, have broad potential utility as well as potential synergism with other targeted agents. In some cases, these agents have demonstrated activity in the form of objective responses.
Ganetespib has some unique properties over all of these agents at the time of their respective licensing deals. Its clinical data package includes hundreds of patients with clinical activity in over five tumor types. Moreover, all of the drugs below were experiencing fierce competition whereas ganetespib is the only Hsp90 inhibitor in advanced clinical development. Lastly, in contrast to the agents below, there are already good biomarkers for patient selection for ganetespib.
For a risk-averse partner, ganetespib’s unique situation creates an attractive and de-risked licensing opportunity. Consequently, investors could expect an upfront payment of ~$40M for the rights in Asia. A worldwide deal should bring upfront payment to over $100M.
With respect to sales potential, Pfizer’s (PFE) crizotinib could be a good benchmark, as a highly effective drug in a niche patient population (ALK fusion mutations, accounting for ~4% of NSCLC cases). According to Bloomberg, crizotinib is forecasted to generate $755M annually in 2015. Hsp90 inhibitors are also believed to be very effective in ALK mutated patients, but there appears to be several additional mutations that make tumors sensitive to Hsp90 inhibition. For example, the patient with the strongest and most durable response in ganetespib’s NSCLC trial appears to have a different mutation (not ALK). Unlike crizotinib, which has so far demonstrated activity in lung and a rare subtype lymphoma, ganetespib already has activity in a wide range of tumors with potentially additional genetic markers for patient selection. Therefore, ganetespib could easily become a $1B drug just in genetically defined patient populations.
Future Catalysts in 2011
Synta is looking at an eventful 2011. Events should include a licensing deal in Asia as well as a large body of clinical data at ASCO 2011 this June. The company is also expected to launch the phase 2b in NSCLC during the quarter, with initial data expected late 2011 or early 2012.
Another important event is first clinical data for CUDC-305, Curis’ (CRIS) Hsp90 inhibitor, partnered with Debiopharm. This could be the first real competitor to ganetespib based on remarks by Curis’ CEO at the recent RBC healthcare conference, who stated that this drug has a clean safety profile and there are multiple signs of clinical activity. One clear advantage CUDC-305 could have is the fact it can be given orally, in contrast to ganetespib. Nevertheless, it is hard to evaluate CUDC-305 as there is no published data for this compound.
In summary, ganetespib has the right ingredients to push Synta’s shares higher during 2011. The drug hits a hot target, it has early signs of clinical activity with an established safety profile and is currently in a broad clinical program. Adding good biomarker data for patient selection and the lack of other advanced stage Hsp90 inhibitors makes ganetespib one of the most attractive unpartnered assets in the industry.
We are adding a second position in Synta in anticipation of multiple catalysts during 2011. In addition, we are selling our positions in Pfizer and Onyx (ONXX) for a 36% and 67% gain, respectively.
Click to enlarge charts