Tecentriq’s chance to seize a poorly served lung cancer type

If the win Roche is touting this morning in Tecentriq’s Impower-131 trial is real it will give the Swiss company an important foothold in a first-line lung cancer setting that is largely untapped by immune checkpoint agents: squamous NSCLC.

However, the magnitude of the progression-free survival in Impower-131 has not been revealed, and it is not clear why an overall survival benefit has, at this interim point, not been seen. But given how poorly served squamous NSCLC is Tecentriq’s chances are good; there is only one other phase III study of an anti-PD-(L)1 MAb specifically in this first-line setting (see table below).

That study is Keynote-407, testing Merck & Co’s Keytruda on top of chemotherapy containing paclitaxel or Abraxane. Keynote-407 has a similar design to Impower-131, and analysts expect it to read out in the first half of this year; however, with an enrolment target of just 560, Keynote-407 is barely half the size of Roche’s rival study.

There are other first-line NSCLC trials under way that recruit subjects with squamous cell histology, including those of Astrazeneca, Pfizer and Regeneron’s respective checkpoint agents Imfinzi, Bavencio and REGN2810. But since these also enrol easier-to-treat non-squamous patients it is not clear whether any benefit in squamous NSCLC could support a label in the latter setting.

At present first-line squamous NSCLC patients have few options beyond Abraxane and carbo-tax chemo; Abraxane, Celgene’s albumin-bound paclitaxel formulation, was approved for first-line squamous and non-squamous NSCLC six years ago. Unlike non-squamous patients, those with a squamous histology cannot get Avastin.

Keytruda monotherapy can be used in the broad NSCLC population as long as patients express PD-L1 above 50%. Opdivo is approved for second-line squamous disease, based on the Checkmate-017 trial, but its failure in the first-line NSCLC study Checkmate-026 cast broader doubts over the Bristol drug’s relevance in lung cancer.

For its part Roche has carefully structured Tecentriq’s front-line NSCLC battle plan around chemo combinations, and Impower-131 is the second phase III trial to read out after the non-squamous Avastin combo study Impower-150.

Both had a three-arm design, and in Impower-131’s case the most relevant comparison was that of Tecentriq plus carboplatin and Abraxane versus carboplatin and Abraxane alone. A third arm, Tecentriq plus carboplatin and unformulated paclitaxel, will be compared against carboplatin/Abraxane control only if the first treatment arm reads out positive for overall survival.

At interim analysis a statistically significant overall survival benefit was not seen, Roche said, and Impower-131 is continuing to full readout unchanged. It is possible that too few subjects have died so far to hit statistical significance, though there might be no survival benefit at all if, for instance, crossover to active treatment has contaminated a potential benefit.

The statistical hit on progression-free survival, a co-primary endpoint of Impower-131, is of course positive, though the magnitude of the benefit must be disclosed to elucidate the clinical relevance; investors will no doubt also want to see the data cut by different levels of PD-L1 expression and other biomarkers. At least the well-known phenomenon of pseudoprogression has not scuppered Impower-131.

At Roche’s full-year financials presentation the company said a positive result in Impower-131 could make it first to market with a PD-(L)1/chemo combo in a setting that accounts for some 30% of the first-line lung cancer market. It is still in with a chance.

This story has been amended, adding two previously omitted Opdivo studies.

To contact the writer of this story email Jacob Plieth in London at [email protected] or follow @JacobPlieth on Twitter