World Lung 2018 – Roche’s chemo combo comes up short

Impower-132 has failed to meet the high bar of Keynote-189, but Roche might still argue that Tecentriq is better than Keytruda in PD-L1-negative patients.

While Roche’s attention in first-line lung cancer remains on getting Tecentriq greenlit in combination with Avastin, full results of the Impower-132 trial were seen as key to enable doctors to make a direct comparison against Merck & Co’s all-powerful approved drug, Keytruda.

Indeed, Roche itself had played up the importance of Impower-132 as being analogous to Merck’s confirmatory Keynote-189 study. On the basis of numerical data, revealed for the first time at the World Congress on Lung Cancer yesterday, it looks like the Swiss firm will have its work cut out, though of course head-to-head comparisons of two separate trials must be carried out with caution.

Both the Merck and Roche studies tested the companies' respective checkpoint antibodies on top of an Alimta/platinum chemo backbone. It was already known that Impower-132 read out positively for progression-free survival, but that a statistically significant overall survival benefit had not been seen at interim analysis.

In Impower-132 Tecentriq reduced risk of progression by 40%, while in Keynote-189 the corresponding risk reduction was a highly impressive 48%.

A cross-trial comparison in 1st-line NSCLC
  Keynote-189 Impower-132
mPFS 8.8mth vs 4.9mth 7.6mth vs 5.2mth
Hazard ratio 0.52 0.60
p value <0.00001 <0.0001
mOS NR vs 11.3mth 18.1mth vs 13.6mth
Hazard ratio 0.49 0.81
p value <0.00001 0.0797
Source: Keytruda label & WCLC 2018. mPFS=median progression-free survival. mOS=median overall survival. NR=not reached.

The fact that the OS benefit, though numerically favouring Tecentriq in Impower-132, has not met statistical significance might be explained by the high number of non-responding subjects who went on to receive second-line immunotherapy.

MD Anderson Cancer Center’s Dr Vassiliki Papadimitrakopoulou, presenting the Impower-132 data at World Lung press yesterday, said 22.4% of control patients went on to receive Bristol-Myers Squibb’s Opdivo, while 9.4% got Keytruda. This data contamination is common in current NSCLC studies, and could raise the question of whether an expensive immunotherapy should be given before chemo.

Strongest case?

Dr Papadimitrakopoulou said the benefit of Tecentriq was seen across key clinical subgroups, and indeed Roche’s strongest case as it tries to square up to Keytruda might be to argue that its drug is more universally active than Merck’s.

Tecentriq’s effect in PD-L1-negative subjects was particularly striking, with a 3.6-month improvement in median PFS amounting to a 55% reduction in risk of progression among this subgroup; 60% of Impower-132 subjects were evaluable for this biomarker. Keynote-189 showed just a one-month PFS increase for PD-L1-negatives given Keytruda, with a 25% risk reduction.

Roche is still awaiting a US FDA decision on approval of Tecentriq plus Avastin in first-line non-squamous NSCLC on the basis of the Impower-150 trial. The regulator is due to give its verdict by December 5, a three-month delay from the original PDUFA date.

Backed by Keynote-189, meanwhile, the Keytruda/Alimta combo secured a full US label on August 20 for non-squamous first-line NSCLC, a use that had initially been granted under accelerated approval based on remissions seen in the small Keynote-021G study. Keynote-189 also led to the Merck drug being greenlit for this indication by the EMA, which had earlier deemed the 021G data insufficient.

Importantly this means that, if patients can tolerate chemo, they can already be prescribed the Keytruda combo without the need to have their tumour assessed for PD-L1 status. Roche might argue that Tecentriq is especially active in PD-L1 negatives, but this might ultimately not matter.

Share This Article