BMY, MRK, and 1st-line NSCLC: Estimating patient shares in the wake of Keynote-189 and Checkmate-227

Print Friendly, PDF & Email


Richard Evans / Scott Hinds

203.901.1631 /.1632

revans@ /


April 22, 2018

BMY, MRK, and 1st-line NSCLC: Estimating patient shares in the wake of Keynote-189 and Checkmate-227

  • After Checkmate-227 top-line results but before Keynote-189 results, we estimated a 20 percent 1st-line NSCLC patient share for Opdivo/Yervoy (BMY). In the wake of both trials’ full results, we estimate an 18 percent share for Opdivo/Yervoy, and a 57 percent share for Keytruda (MRK). The net effect of the two trials is a significant expansion of PD-1-based first-line treatment in NSCLC, rather than a zero-sum shift of patient share from one regimen to the other
  • Further details regarding tumor mutational burden (TMB) and PD-L1 sub-groups, and/or duration of response, could shift our estimates either way, though we suspect duration of response is more likely to advantage Opdivo/Yervoy
  • These two trials are less about which PD-1 inhibitor is best, and more about whether adding chemotherapy or a CTLA4 to a PD-1 inhibitor is best. The answer appears to be that it depends on histology, and of course PD-L1 and TMB status. Looking forward, and for the moment ignoring the potential addition of 3rd generation immune-based mechanisms, the question seems to be whether combined therapy with a PD-1 inhibitor, chemotherapy, and CTLA4 is the right way to go. To our knowledge only BMY is evaluating this approach, in Checkmate-9LA, whose schedule primary completion date is August 2019


MRK’s Keynote-189 trial shows that Keytruda in combination with chemotherapy is superior to chemotherapy alone in first-line NSCLC patients whose tumors lack a driver mutation and have non-squamous histology. BMY’s Checkmate-227 trial shows that Opdivo+Yervoy is superior to chemotherapy in first-line NSCLC patients whose tumors lack a driver mutation and have high tumor mutational burden (TMB), regardless of histology

Keynote-189 proves the Keytruda/chemo combination’s superiority to chemotherapy alone in terms of both overall survival (OS) and 12-month progression free survival (PFS). Checkmate-227 proves the Opdivo/Yervoy combination’s superiority to chemotherapy on the basis of 12-month PFS alone. OS is the gold standard, which all else equal gives the Keytruda/chemotherapy combination an advantage over Opdivo/Yervoy simply in terms of the weight of its evidence. However, because PFS is the only primary endpoint shared by the two trials, it’s the only practical basis for comparison of the two trials’ results

In Keynote-189, the hazard ratio (HR) for PFS between the Keytruda/chemotherapy and chemotherapy arms was 0.52 in the combination’s favor, with a 95 percent CI of 0.43 – 0.64. In Checkmate-227, among all patients (irrespective of PD-L1 or TMB levels), the HR for PFS was 0.83 in Opdivo/Yervoy’s favor, with a 95 percent CI of 0.72 – 0.96. Among patients with high TMB, regardless of PD-L1 level, the HR for PFS was 0.58, with a 95 percent CI of 0.43 – 0.77. And finally, among patients with non-squamous histology and high TMB, regardless of PD-L1, the HR for PFS was 0.55, with a 95 percent CI of 0.38 – 0.80

Our objective here is to estimate which first-line, non-driver patients go to which of the two regimens. Starting with histology, we expect patients with squamous tumors to be divided first by PD-L1 expression level, and then by TMB status. Keynote-189 excluded patients with squamous histology; however, in Keynote-24 Keytruda alone showed significantly better 12-month PFS than chemotherapy in patients with squamous tumors and PD-L1 expression of ≥ 50% (HR 0.35, 95 percent CI: 0.17 -0.71) (Exhibit 1, A). For patients with PD-L1 < 50% we expect those with high TMB (Exhibit 1, B) will be treated with Opdivo/Yervoy, and that those with low TMB (Exhibit 1, C) will be treated with standard chemotherapy. In Checkmate-227 Opdivo/Yervoy was directionally but not statistically significantly better than standard chemotherapy in patients with squamous tumors and high TMB, regardless of PD-L1 expression (HR 0.63, CI: 0.39 – 1.04)

Shifting to patients with non-squamous tumors, we expect PD-L1 non-expressors (<1%) with high TMB to go to Opdivo/Yervoy (Exhibit 1, D), and PD-L1 non-expressors with low TMB to go to the Keytruda combination (Exhibit 1, E). In Keynote-189 the HR for PFS in <1% PD-L1 expressors was 0.75 in Keytruda’s favor, but with a 95 percent CI of 0.53 – 1.05. In Checkmate-227 the HR for PFS in patients with <1% PD-L1 and high TMB was 0.48, with a 95 percent CI of 0.27 – 0.85

Staying with non-squamous tumors, we expect PD-L1 expressors (≥1%) with low TMB to go to the Keytruda combination (Exhibit 1, F), and expect PD-L1 expressors with high TMB to be divided among the Keytruda combination and Opdivo/Yervoy (Exhibit 1, G), with the Keytruda combination being favored on the basis of weight of evidence, since the Keytruda combination has proven an OS advantage but Opdivo/Yervoy has not. In Checkmate-227 the Opdivo/Yervoy HR for PFS in non-squamous tumors with high TMB, irrespective of PD-L1 expression, was 0.55 (95 percent CI: 0.41 – 0.81); across all non-squamous patients irrespective of TMB or PD-L1 the Keytruda/chemo HR for PFS was 0.52 (95 percent CI: 0.43 – 0.64). Thus in the case of a non-driver / non-squamous patient with high TMB, Opdivo/Yervoy’s HR for PFS is comparable (0.55 v. Keytruda/chemo’s 0.52), and the confidence intervals overlap

Summarizing the patient allocations in Exhibit 1; Keytruda or the Keytruda/chemotherapy combination captures at least 43 percent of patients[1], plus some share of the 18 percent of first-line NSCLC patients who are non-squamous PD-L1 expressors with high TMB. If we reasonably assume the Keytruda/chemotherapy combination captures 75 percent of these patients, this implies a total first-line NSCLC patient share for Keytruda and the Keytruda/chemotherapy combination of 57 percent. The Opdivo/Yervoy combination should capture at least 13 percent of first-line NSCLC[2], plus an assumed one-quarter of the 18 percent of first-line NSCLC patients who are non-squamous PD-L1 expressors with high TMB, for a total assumed patient share of 18 percent. Note that this share estimate for Opdivo/Yervoy compares favorably with the 20 percent share estimate we provided after BMY reported top-line Checkmate-227 results, and before MRK reported Keynote-189[3]

Two things still might change the takeaways from these two pivotal trials, specifically any further characterization of responses by TMB in Keynote-189 or by PD-L1 in Checkmate-227, and finalization of duration of response data from both trials

Based on these trials’ evidence, plus evidence beyond these two trials that indicates PD-L1 and TMB are similarly powerful but independent predictors of response to PD-1 inhibitors, we’re assuming Keytruda/chemotherapy is favored when TMB is low (regardless of PD-L1), and that Opdivo/Yervoy is favored when TMB is high and PD-L1 is not expressed. More granular disclosure of responses by combined PD-L1/TMB status from either trial could conceivably shift the balance in favor of one regimen or the other

In patients where the current PFS evidence supports either regimen (i.e. the 18 percent of patients who are non-squamous PD-L1 expressors with high TMB), the Keytruda/chemotherapy combination has the advantage of having proven overall survival (OS). We know from these trials that chemotherapy is additive to Keytruda, and that Yervoy is additive to Opdivo. And we know from earlier trials that duration of response to chemotherapy generally is less than duration of response to Yervoy. It’s too early to reach conclusions about relative response durations across the two trials; however it’s entirely possible that duration of response eventually could favor Opdivo/Yervoy in this significant block of patients

And finally, if we reasonably assume that Keytruda and Opdivo are similarly effective as PD-1 inhibitors, something the broader clinical evidence certainly supports, then we can think of the Keynote-189 v. Checkmate-227 comparison largely in terms of a comparison of PD-1+chemotherapy to PD-1+CTLA4. This leads to the obvious question of whether PD-1+chemotherapy+CTLA4 ‘triple therapy’ is better than either of the ‘dual therapies’ tested in these two trials. To our knowledge only BMY is evaluating this triple combination; Checkmate-9LA (NCT03215706) is a phase 3 evaluation of Opdivo/Yervoy/chemotherapy v. chemotherapy alone in first-line NSCLC in non-driver patients, regardless of histology, PD-L1, or TMB. The trial has overall survival (OS) as a primary endpoint, and a scheduled primary completion date of August 2019


  1. Squamous with PD-L1 ≥ 50%, non-squamous PD-L1 non-expressors with low TMB, and non-squamous PD-L1 expressors with low TMB
  2. Squamous with PD-L1 <50% and high TMB, and non-squamous PD-L1 non-expressors with high TMB
  3. “BMY/MRK: Why CM227 is very good news for BMY, and not such bad news for MRK”, SSR Health LLC, February 8, 2018; pg8, Exh 7


©2018, SSR, LLC, 225 High Ridge Rd, 2nd Floor, Stamford, CT 06905. All rights reserved. The information contained in this report has been obtained from sources believed to be reliable, and its accuracy and completeness is not guaranteed. No representation or warranty, express or implied, is made as to the fairness, accuracy, completeness or correctness of the information and opinions contained herein. The views and other information provided are subject to change without notice. This report is issued without regard to the specific investment objectives, financial situation or particular needs of any specific recipient and is not construed as a solicitation or an offer to buy or sell any securities or related financial instruments. Past performance is not necessarily a guide to future results. In the past 12 months, through a wholly-owned subsidiary SSR Health LLC has provided paid advisory services to BioPharmX (BPMX), Pfizer Inc (PFE), Gilead Sciences (GILD), Bristol-Myers Squibb (BMY) and Sanofi (SNY) on both securities-related and non-securities-related topics. One or more of SSR Health’s analysts owns long positions in the following stocks: ACOR, AGEN, AGIO, AKAO, ALKS, ALNY, ANAB, ARRY, BMY, CPRX, DERM, DPLO, DVA, ESALY, ESRX, FOLD, GWPH, INCY, INSM, IONS, ITCI, KALA, LOXO, LXRX, MDCO, NSTG, PGNX, PRTK, PTLA, RHHBY, SRPT, STML, TBPH, TRVN, TSRO, TTPH, TXMD, VRTX, VSTM

Print Friendly, PDF & Email