The landscape for first-line metastatic NSCLC without driver mutations has been shifting constantly over the last few months, with a truly dizzying cadence of new clinical data releases for immune checkpoint inhibitors as single agents and as part of combination regimens. The 2018 ASCO meeting was no exception to this trend with NSCLC taking central stage and a wealth of new data presented for Keytruda (KEYNOTE-407 and KEYNOTE-042), Tecentriq (IMpower150 and IMpower131), and Opdivo (CheckMate-227) in first-line metastatic NSCLC.
The table below summarizes some of the key data from these trials, and we’ve also included data from KEYNOTE-024 and KEYNOTE-189 to provide additional—but essential—context. We then discuss some of the key take-home messages from these trials and how they could affect the future landscape for immune checkpoint inhibitors in the first line.
Select Phase III Results for Immune Checkpoint Inhibitors in First-Line Metastatic NSCLC
Keytruda improves OS for all patients with at least 1% tumor PD-L1 expression.
- Data from KEYNOTE-042 were presented at the plenary session and showed that patients receiving Keytruda alone lived a median 4.6 to 7.8 months longer than those who received chemotherapy across the three pre-specified thresholds for tumor PD-L1 expression.
- These findings not only confirm the OS benefit seen in the KEYNOTE-024 trial (which looked at patients with PD-L1 ≥50%, and where OS was a secondary end point) but also demonstrate that the OS benefit extends to all patients with at least 1% PD-L1 expression. Perhaps more importantly, Keytruda was much better tolerated than chemotherapy, making it a suitable option for patients who are ineligible for (or who decline) chemotherapy.
Merck outshines Roche in combinations for squamous NSCLC.
- Until recently, the benefit of adding an immune checkpoint inhibitor to chemotherapy had only been demonstrated in nonsquamous NSCLC (Keytruda + chemotherapy in KEYNOTE-189). Results from KEYNOTE-407 show that Keytruda + chemotherapy extends OS and PFS versus chemotherapy alone in patients with squamous NSCLC irrespective of PD-L1 expression status. These results are practice-changing and mark another strong win for Merck in this population, on the back of the KEYNOTE-024 monotherapy data that also included squamous patients.
- Data for Tecentriq + chemotherapy appear less impressive in this context. In IMpower131, Tecentriq + chemotherapy significantly extended PFS (co-primary end point) with a doubling of the 12-month PFS rate (25% vs 12%) but the first interim OS analysis did not suggest any improvement in OS with the combination (the next interim OS analysis is expected later in 2018).
Several positive trials for chemoimmuno-combinations in nonsquamous NSCLC
- Interim results of IMpower150 suggest that adding Tecentriq to Avastin + chemotherapy improves MOS over Avastin + chemotherapy in nonsquamous patients, but that substituting Avastin with Tecentriq as an add-on to chemotherapy does not (although data are immature for this latter comparison as the efficacy boundary has not yet been crossed).
- The positive KEYNOTE-189 data for Keytruda plus chemotherapy published a couple of weeks prior to ASCO 2018 arguably set this regimen as a “truer” comparator in the nonsquamous setting. While cross-trial comparisons are risky, we note that the OS HR is more favorable for Keytruda + chemotherapy than Tecentriq + Avastin + chemotherapy, giving the former combination a possible edge at this point.
- Crucially, IMpower150 did allow the inclusion of EGFR+/ALK+ patients after prior TKI therapy (unlike the other competing trials) and data in this subset suggest that Tecentriq + Avastin + chemotherapy prolongs OS (data not shown).7 This finding could help differentiate Tecentriq with its competitors in previously treated patients with driver mutations, for whom the evidence to date had pointed to a limited role for immune checkpoint inhibitors.
- Of note, Roche reported in May 2018 that the Phase III IMpower130 trial of Tecentriq + chemotherapy (carboplatin and Abraxane) in first-line metastatic nonsquamous NSCLC met its co-primary end points of OS and PFS, with data to be unveiled in a future medical meeting9 (possibly ESMO 2018). Results from this trial will shed more light on the role of Tecentriq-based combinations and how they compare with other immunotherapy combinations in the nonsquamous setting.
High TMB may be prognostic for Opdivo plus Yervoy or plus chemotherapy but more robust prospective data are needed
- Tumor mutation burden (TMB) testing in CheckMate-227 is a strategy that was not specified in the initial trial design but one that occurred after the study had started, which somewhat lowers the level of confidence in these data versus a truly prospective trial. We already knew before ASCO 2018 that Opdivo + Yervoy had met the co-primary end point of PFS in patients with high TMB (defined as ≥10 mutations/megabase) irrespective of PD-L1 expression level or histology.10
- Data shown at ASCO 2018 suggest that TMB and PD-L1 have more predictive value together—adding Opdivo to Yervoy or to chemotherapy leads to better PFS than chemotherapy alone in the subset of PD-L1 <1% patients who have high TMB specifically. Opdivo + Yervoy was safer and led to more durable responses than Opdivo + chemotherapy in this subset (data not shown). While these results are intriguing, the all-important OS data have not yet been reported.
- There are question marks over the feasibility of TMB testing in routine practice, in part because of the relatively long turnaround time (a few weeks, versus days for IHC testing) and the amount of tissue required. Only 48-58% of patients in CheckMate-227 were evaluable for TMB; this percentage may be even lower in real clinical practice and prove to be a barrier to uptake of Opdivo-based combinations if approved.
What are the market and clinical practice implications of these data?
With so many trials reporting at the same time, it is inevitable that physicians will make indirect comparisons between them to help in decision making (with all the caveats of cross-trial comparisons). Several of these trials in the first-line setting have demonstrated OS benefits over comparators – in this context, any new immunotherapy (single-agent or combination) will need to demonstrate a clinically significant improvement in OS if it is to get a slice of the first-line metastatic NSCLC market.
The powerful narrative of four positive trials for Keytruda in first-line metastatic NSCLC is likely to consolidate Merck as the leader in the large and commercially lucrative first-line metastatic setting, ahead of Bristol-Myers Squibb and Roche should their combinations gain approvals. This also means that the stakes are now even higher for AstraZeneca and its Phase III MYSTIC trial for Imfinzi ± tremelimumab, which did not meet PFS and for which the final OS readout is expected in the second half of 2018.
It is still unclear which patients would benefit the most from single-agent versus combination approach—not least because of the lack of head-to-head studies—and there’s plenty of room for debate, but we can start brushing the following picture at this point:
- It is likely that Keytruda monotherapy will continue to dominate in patients with PD-L1 ≥50% (irrespective of histology) when balancing efficacy and safety compared with that of Keytruda + chemotherapy regimens.
- Keytruda monotherapy may be a more debatable choice for patients with 1-49% PD-L1 expression, partly because the KEYNOTE-042 data suggest that the benefit seen with Keytruda alone in PD-L1 ≥1% was likely driven by the subset of PD-L1 ≥50% patients (which was about half of the total trial population). Patients with 1-49% PD-L1 expression may benefit more from a combination therapy, such as Keytruda + chemotherapy.
- The optimal choice of treatment for patients with PD-L1 <1% is more difficult still. While Keytruda + chemotherapy is an option, patients with high TMB may also benefit from Opdivo-based regimens. Patients with low PD-L1 and low TMB may well be suited to receive chemotherapy alone.
DRG’s coverage of NSCLC is extensive, including primary market research reports on U.S. Treatment Sequencing (just published) and U.S. Current Treatment: Physician Insights (publishing this month). We’ll update the market forecast for NSCLC with a 2017 base-year in Q4 2018 (see the current Disease and Landscape forecast here).
- Brahmer J, et al. Updated analysis of KEYNOTE-024: pembrolizumab vs platinum-based chemotherapy for advanced NSCLC with PD-L1 TPS ≥50%. 2017 WCLC meeting. Abstract OA 17.06.
- Merck & Co, press release, October 18, 2017.
- Keytruda prescribing information, US FDA, November 2017.
- Lopes G, et al. Pembrolizumab (pembro) versus platinum-based chemotherapy (chemo) as first-line therapy for advanced/metastatic NSCLC with a PD-L1 tumor proportion score (TPS) ≥ 1%: open-label, Phase 3 KEYNOTE-042 study. 2018 ASCO meeting. Abstract LBA4.
- Gandhi L, et al. Pembrolizumab plus chemotherapy in metastatic non-small-cell lung cancer. NEJM. 2018;378:2078-2092.
- Paz-Ares LG, et al. Phase 3 study of carboplatin-paclitaxel/nab-paclitaxel (Chemo) with or without pembrolizumab (Pembro) for patients (Pts) with metastatic squamous (Sq) non-small cell lung cancer (NSCLC). 2018 ASCO meeting. Abstract 105.
- Socinski MA, et al. Overall survival (OS) analysis of IMpower150, a randomized Ph 3 study of atezolizumab (atezo) + chemotherapy (chemo) ± bevacizumab (bev) vs chemo + bev in 1L nonsquamous (NSQ) NSCLC. 2018 ASCO meeting. Abstract 9002.
- Jotte RM, et al. IMpower131: primary PFS and safety analysis of a randomized Phase III study of atezolizumab + carboplatin + paclitaxel or nab-paclitaxel vs carboplatin + nab-paclitaxel as 1L therapy in advanced squamous NSCLC. 2018 ASCO meeting. Abstract LBA9000.
- Roche, press release, May 29, 2018.
- Hellmann MD, et al. Nivolumab plus ipilimumab in lung cancer with a high tumor mutational burden. NEJM. 2018; 378:2093-2104.