The much-anticipated latest installment of the first-line non-small cell lung cancer (NSCLC) battle took place at this year’s American Society of Clinical Oncology (ASCO) annual meeting. Roche, Bristol Myers Squibb (BMS) and Merck presented data in various subsets of patients, but Merck remains the current winner in this large and lucrative population.
For any new immuno-oncology combination to be adopted as standard of care (SOC) in first-line NSCLC, it must show overall survival (OS) benefit, since this has already been achieved in several studies, including KEYNOTE-024 (Keytruda vs. platinum doublet in patients with ≥50% PD-L1 expression), and KEYNOTE-189 (Keytruda + carboplatin/pemetrexed vs. chemotherapy alone in non-squamous patients, irrespective of PD-L1). Key abstracts at ASCO 2018 included:
- Roche’s IMpower131 study of Tecentriq with carboplatin and nab-paclitaxel in squamous NSCLC (LBA9000)
- Merck’s KEYNOTE-407 study of Keytruda plus chemotherapy in squamous NSCLC (105)
- BMS’s CheckMate 227 trial of Opdivo + Yervoy in squamous or non-squamous patients with high tumor mutational burden (TMB) (9001)
- Merck’s KEYNOTE-042 trial of Keytruda monotherapy versus platinum chemotherapy in PD-L1-positive patients with any histology (LBA4)
- Roche’s IMpower150 study of carboplatin, paclitaxel, Avastin and Tecentriq in non-squamous patients (9002)
Although Roche posted robust data from IMpower131, Merck’s Keytruda is still the checkpoint agent of choice in patients with squamous histology. In the intent-to-treat (ITT) population, unstratified by PD-L1 status, Tecentriq plus chemotherapy achieved progression-free survival (PFS) but not OS benefit versus chemotherapy alone (median PFS hazard ratio (HR) 0.71, p=0.0001, median OS HR 0.96, p=0.6931).
However, this was bested by Keytruda in KEYNOTE-407 in the same population, which showed both PFS and OS benefit in combination with chemotherapy compared to chemotherapy alone (median PFS HR 0.56, p<0.0001, median OS HR 0.64, p=0.0008). In addition, despite benefit being seen in PFS (HR 0.44) and OS (HR 0.56) in the PD-L1 high population in IMpower131, clinically this is the same population that would be eligible for Keytruda monotherapy, which clinicians are likely to prefer due to the lower toxicity. This was reflected by the expert discussant in one of the NSCLC sessions at ASCO, Dr Thomas Stinchcombe, who included Keytruda in his current preferences for first-line squamous NSCLC treatment patients, in all PD-L1 subgroups: Keytruda monotherapy for PD-L1 ≥50% (KEYNOTE-024), and Keytruda plus carboplatin/paclitaxel for PD-L1 1-49% or <1% (KEYNOTE-407).
Keytruda also preferred for squamous patients
For squamous patients, Keytruda also remains the preferred checkpoint agent, despite several new datasets at ASCO incorporating this subgroup. BMS presented data from CheckMate 227 in patients with <1% PD-L1 expression of any histology, and also discussed stratification by TMB.
It was reported at AACR 2018 that the Opdivo + Yervoy arm had met its co-primary endpoint of improvement in PFS versus platinum doublet chemotherapy for patients with high TMB, ≥10 mut/Mb (PFS HR 0.58). At ASCO, PFS benefit versus chemotherapy (HR 0.74) in patients with <1% PD-L1 was disclosed, but this was deemed “pedestrian”, and will not be sufficient to change SOC.
Retrospective analysis of TMB subgroups in both Opdivo arms was more interesting: in patients with <1% PD-L1 expression, PFS benefit with both Opdivo plus chemotherapy, and Opdivo plus Yervoy was increased in the high TMB (≥10 mut/Mb) subgroup (Opdivo plus chemotherapy PFS HR 0.56, Opdivo plus Yervoy PFS HR 0.48). However, until prospective data is available correlating this directly with survival, these data will also not change SOC.
Merck’s KEYNOTE-042 study aimed to show benefit from Keytruda beyond the ≥50% PD-L1 cut-off already demonstrated in KEYNOTE-024. However, while OS benefit was significant in all pre-planned PD-L1 groups (≥50% HR 0.69, ≥20% HR 0.77, ≥1% HR 0.81), the exploratory analysis of OS in the 1-49% PD-L1 subgroup only showed a HR of 0.92. The secondary endpoint of PFS was also not met in the ≥50% PD-L1 subgroup (HR 0.81, p=0.017, did not cross significance threshold), meaning that statistical analysis was not performed in the lower PD-L1 expression subgroups, per protocol. This will be re-assessed at a later time-point, but for now this study meets ASCO guidelines for meaningful clinical trial outcomes only for squamous patients, and chemotherapy will likely continue to be used for non-squamous patients with low PD-L1 expression.
Tecentriq to lose out?
While it is clear that OS benefit with Keytruda monotherapy is driven by higher PD-L1 expression, given the positive OS data in patients ≥1% PD-L1 expression, these results are expected to lead to label expansion beyond the current PD-L1 ≥50% limit. However, better combination strategies may provide alternative options for these patients in the future, leaving Keytruda’s superiority as a monotherapy in this population less secure than in other subgroups.
OS data from IMpower150 was clearly positive across all PD-L1 subgroups (ITT HR 0.78, p=0.0164), however in non-squamous ALK- or EGFR-wild-type patients, the incremental effect compared to previously reported OS from KEYNOTE-189 was shown by the discussant in this session to be statistically inferior (KEYNOTE-189 OS HR 0.49, p<0.00001). Unfortunately for Roche, this means that Tecentriq may lose out to Keytruda once more.
Stinchcombe’s preferred treatment options for non-squamous patients again put Merck in the driver’s seat: Keytruda monotherapy for PD-L1 ≥50% (KEYNOTE-024), Keytruda plus carboplatin/pemetrexed for PD-L1 1-49% (KEYNOTE-189), and either chemotherapy alone or Keytruda plus carboplatin/pemetrexed for PD-L1 <1% (KEYNOTE-189), depending on TMB status. Interestingly, the most uncertainty remains about the benefit of adding immunotherapy to SOC in this low-PD-L1 non-squamous population.
The treatment landscape in first-line NSCLC is increasingly segmented and rapidly evolving, and while these data at ASCO bring physicians closer to clear-cut decisions on how to best treat patients, unanswered questions remain.
GlobalData (2017). Non-Small Cell Lung Cancer (NSCLC): Dynamic Market Forecast to 2025, August 2017, GDHC001FS.
For more insight and data, visit the GlobalData Report Store – Pharmaceutical Technology is part of GlobalData Plc.