Roche has shared phase 3 data on the drug that enhanced the efficacy of Johnson & Johnson’s Zytiga in a subset of prostate cancer patients earlier this year. The trial squeezed under the bar for statistical significance by improving on the median progression-free survival (PFS) of Zytiga by two months.
In June, Roche revealed adding the small molecule Akt inhibitor ipatasertib to Zytiga had improved on the PFS achieved by J&J’s blockbuster alone in a genetic subset of prostate cancer patients. The combination failed to move the needle in the broader population, suggesting use of Roche’s drug may be limited to the roughly 50% of metastatic castrate-resistant prostate cancer patients who have PTEN-loss tumors.
At the ESMO Virtual Congress 2020, Roche shared the numbers behind the earlier top-line findings. In patients with PTEN-loss tumors by immunohistochemistry, the median PFS of the combination was 18.5 months, compared to 16.5 months for Zytiga alone. That difference translated into a p value of 0.0335, resulting in the trial meeting one of its co-primary endpoints. The hazard ratio was 0.77. It is unclear where that leaves ipatasertib in prostate cancer.
“Although initial data are encouraging, OS benefit and additional secondary endpoints are not yet mature. The trial will continue until the next planned analysis, and data will be shared and discussed with health authorities,” a spokesperson for Roche’s Genentech said.
A bigger difference between the PFS of the arms was seen in a subset of patients who had PTEN-loss determined by next-generation sequencing. In those patients, the median PFS in the combination arm was 19.1 months, versus 14.2 months in the control group. That analysis included less than one-fifth of the overall 1,101-subject study population, though.
Roche stratified patients by PTEN status in light of the inhibitory role the tumor suppressor plays in the PI3K/Akt survival pathway targeted by ipatasertib. Patients with PTEN-loss tumors have activated PI3K/Akt survival pathways, driving cancer growth and pointing to the therapeutic potential of Akt inhibitors in patients with the mutation.
The top-line phase 3 finding that ipatasertib improved median PFS in patients with PTEN-loss tumors but not the broader trial population appeared to back up the hypothesis that Akt inhibitors will work best in the subpopulation. However, the data behind the top-line findings are more equivocal.
In the broader intention-to-treat (ITT) population, the ipatasertib-Zytiga combination had a median PFS of 19.2 months, numerically longer than the result in PTEN-loss patients. However, with the control arm recording a median PFS of 16.6 months, the combination fared worse statistically in the wider study population. The p value was 0.0431 and the hazard ratio was 0.84.
A similar picture emerged in the secondary endpoints, with ipatasertib performing comparably in the PTEN-loss patients and wider population. The median times to PSA progression in the PTEN and ITT analyses were 12.6 months and 12.9 months, respectively. The control arm was around 8 months in each population. Ipatasertib achieved an identical response rate—61%—in both populations.
If use of ipatasertib is limited to patients with PTEN-loss tumors, Roche will need to get more people tested for the mutation to make the drug a commercial success. PTEN tests exist but aren’t part of the standard response to a prostate cancer diagnosis.
Roche used ESMO to share data from another study that gave ipatasertib and paclitaxel to a genetic subset of breast cancer patients. The addition of ipatasertib had no apparent effect, with the control and treatment arms achieving identical median PFS, response rates and response durations.
Analysts at Jefferies called the breast cancer results “disappointing,” adding that they raise questions “about whether patients with PIK3CA/AKT1/PTEN-altered tumours are the best target population for ipatasertib.”