A very happy New Year everyone! After shaking off the dust of an extended break over the last two weeks, this morning brought plenty of news to kick start 2012.
The most interesting news was AVEO-Astellas’ announcement regarding their VEGF inhibitor, tivozanib, in advanced renal cell cancer (RCC):
“Tivozanib demonstrated superiority over sorafenib in the primary endpoint of progression-free survival (PFS) in TIVO-1, a global, randomized Phase 3 clinical trial evaluating the efficacy and safety of investigational drug tivozanib compared to sorafenib in 517 patients with advanced renal cell carcinoma (RCC).”
The previous front-line trials with sorafenib (Nexavar) and sunitinib (Sutent) were randomised trials that compared the active drug to placebo and interferon respectively, so this is the first head to head study that compares an investigational agent to an approved VEGF therapy. It was a risky study given that the prior phase II trial in RCC compared tivozanib against placebo, meaning AVEO had no idea whether their had a superior agent or not.
What did the data show?
The TIVO-1 study sought to compare tivozanib versus sorafenib in patients with clear cell renal carcinoma who were treatment naïve, ie they had not received prior therapy with either a VEGF or an mTOR inhibitor. The companies announced the topline results from the DSMC analysis:
- Tivozanib demonstrated a statistically significant improvement in PFS with a median PFS of 11.9 months compared to a median PFS of 9.1 months for sorafenib in the overall study population.
- Tivozanib demonstrated a statistically significant improvement in PFS with a median PFS of 12.7 months compared to a median PFS of 9.1 months for sorafenib in the pre-specified subpopulation of patients who were treatment naïve (no prior systemic anti-cancer therapy); this subpopulation was approximately 70% of the total study population.
- Tivozanib demonstrated a well-tolerated safety profile consistent with the Phase 2 experience; the most commonly reported side effect was hypertension, a well established on-target and manageable effect of VEGFR inhibitors.
The data is being submitted as an abstract to ASCO and an FDA filing will also most likely result this year.
No further information was available at this stage on the second point for the treatment-naive and prior therapy groups. The safety profile was reported on the call to be in-line with prior phase II experience, meaning hypertension, a VEGF class effect, was the most common adverse event.
What do these data mean?
There are several points to note from this topline analysis:
- The sorafenib arm did better than expected – based on the PI, we would expect around 5.5 months not 9.1 months.
- The sunitinib front-line trial showed a median benefit of ~11.0 months, based on the PI giving 47.3 weeks for PFS.
- So while the difference in the TIVO-1 study of ~3 months between the two arms is smaller, the absolute PFS benefit for tivozanib is slightly better than sunitinib, although the caveat is that we cannot really compare them clinically without a head to head trial.
- We don’t yet know how well Pfizer’s axitinib (Inlyta) will do in the 1st line setting, as only the second-line topline findings were announced in November:
“Axitinib in previously treated patients with metastatic renal cell carcinoma (mRCC), has met its primary endpoint, demonstrating that axitinib significantly extended progression-free survival (PFS) when compared to sorafenib, in the study population.”
My guess is that at this rate, tivozanib and axitinib may well prove to be more similar than different, as the compounds were fairly close preclinically. By this, I mean that with more selective VEGF inhibitors we are likely to see less toxicities from off-kinase inhibition with perhaps similar efficacy to Sutent, than vast improvements in PFS.
Ultimately, the real winner in a very crowded RCC market is going to be a drug with not only superior PFS, but also a cleaner safety profile. It has been shown previously that combining sunitinib with an mTOR such as everolimus (Afinitor) led to unacceptable toxicities and the concept was thereby abandoned after the phase I trial. However, if either tivozanib or axitinib can be safely combined with an mTOR for additional efficacy, then this would be a big plus for both physicians and patients. At the moment, the count is out on that option.
Currently, physicians typically treat patients with Sutent initially and then consider Nexavar, Afinitor or Torisel as second-line therapy, while reserving Votrient (pazopanib), another VEGF inhibitor, for 3rd line given its adverse event and efficacy profile. Sequencing multiple single agents is therefore very much derigeur in RCC. Whether tivozanib or axitinib can change this is an important consideration for the future.
And finally, some observations…
This year’s ASCO is going to be a most interesting meeting for renal cell cancer with data now expected for tivozanib in first and second line expected versus sorafenib, following on from the head to head of axitinib versus sorafenib in second line at last year’s ASCO. I’m not sure when the axitinib front-line data will be available, but it is eagerly awaited given the tivozanib data announcement.
It is interesting that the market is down on the news with AVEO’s shares falling, despite meeting the primary study endpoint. No doubt investors were expecting either a bigger difference between the arms or for the absolute overall PFS value for tivozanib to be greater than 12-12.5 months in the upfront setting.
Part of me would have liked to have seen the safety and efficacy signal of the head to head with sorafenib in a phase 2 trial to see what the real difference between the two was earlier. As it is, the phase 2 was conducted versus placebo, which doesn’t tell us anything about what might be expected in a phase 3 study. It’s always a risk switching horses into the unknown.
Mechanistically though, tivozanib is a VEGF inhibitor in the vein as Sutent, Nexavar, Avastin and Votrient, albeit a more selective one, so there is a limit on what can be expected by blocking a single pathway. Resistance inevitably sets in, so going forward we need to determine what the adaptive resistance pathways to therapy are and then determine what combinations might be feasible to try and overcome them. If this can be done without excessive toxicities, then we can expect the survival to improve further.
Another thing to consider is that previously, AVEO presented some rather promising data on biomarkers of response in their preclinical and phase I RCC data. I’m very keen to see if any subset analysis in the phase III trial will be presented at ASCO and whether the hypothesis stands up in the clinical setting for either the treatment naïve or relapsed patients. If the biomarker can determine who is most likely to respond to tivazanib, then I would expect the PFS to improve in that subset. We must wait and see what, if anything, happens with that.
Finally, for patients moving from 10 to almost 12 months may sound like a small increment in survival, but when we consider the cumulative effect of sequencing multiple drugs, there is no doubt that the efficacy benefits begin to accrue over time. It should not be forgotten than until recently, the choices in RCC were stark. There was either IL-2 (only suitable for a limited number of patients due to severe toxicities) or interferon, which typically had a PFS of approx. 4-5 months.
Since then, in only a few years with several new therapies entering the market, the PFS has already doubled with only single agent therapy; that’s quite an improvement, although we still have a long way to go in understanding the biology of the disease.