BMY: Why Opdivo still may dominate NSCLC

Print Friendly, PDF & Email

Richard Evans / Scott Hinds / Ryan Baum / Hardy Evans


203.901.1631 /.1632 / .1627

revans@ / shinds@ / rbaum@ /

August 8, 2016

BMY: Why Opdivo still may dominate NSCLC

  • BMY’s PD-1 inhibitor Opdivo failed to meet its primary endpoint (progression free survival) v. standard chemotherapy in treatment-naïve non-small cell lung cancer (NSCLC) patients, soon after MRK’s PD-1 Keytruda had succeeded in a similar comparison
  • However, the Opdivo trial (Checkmate-026) included patients with PDL1 expression levels of 5% or greater, as compared to the Keytruda trial (Keynote-024) which limited enrollment to patients with much greater (≥ 50%) PDL1 expression levels
  • Opdivo’s current NSCLC label is for 2nd-line use (after failure of platinum-based chemotherapy) in patients irrespective of PDL1 levels; conversely Keytruda’s current NSCLC label (also for 2nd-line use, but soon to be for 1st-line use) is restricted to patients with higher PDL1 expression
  • The trials on which Opdivo’s current label is based showed mixed signals with respect to the relationship between PDL1 expression and efficacy; however, looking across the entirety of Opdivo’s NSCLC evidence, including in particular the phase 1b precursor (Checkmate-012) to Checkmate-026, it’s clear that Opdivo almost certainly is more effective in tumors with higher PDL1 expression, and vice versa. As such the inclusion of patients with low PDL1 levels in Checkmate-026 likely accounts for Opdivo’s failure
  • The full Checkmate-026 analysis may yet give Opdivo an advantage over Keytruda. In all likelihood, patients whose PDL1 expression levels fall just below the 50% Keytruda threshold can still benefit from PD-1 based agents – and Checkmate-026 is reasonably likely to prove that Opdivo is effective in these patients (as well as in patients above the 50% threshold). Importantly, only about 25% of NSCLC patients have PDL1 expression levels of 50% or higher, so the majority of the NSCLC market lies below the Keytruda threshold
  • Even if Opdivo completely loses the 1st-line, single-agent, high-PDL1 segment of the NSCLC market to Keytruda – which we see as unlikely – Opdivo still has a very good chance of dominating elsewhere in NSCLC. BMY’s NSCLC strategy evaluates Opdivo in combination with either a CTLA4 (Yervoy) or in combination with standard chemotherapy; in contrast, MRK’s NSCLC strategy (for patients with lower PDL1 levels) is more narrowly focused on Keytruda in combination with standard chemotherapy. In phase 1b BMY’s Opdivo + Yervoy combination has shown very high response rates (57%) as 1st-line NSCLC therapy across all patients regardless of PDL1 expression levels, and even higher (92%, 12/13) response in patients with PDL1 expression levels of 50% or greater
  • NSCLC accounts for 21% of the mortality associated with cancers for which CI’s are either approved, or being evaluated in major trials; NSCLC patients with ≥ 50% PDL1 expression likely account for roughly 5% of total mortality for these cancers. BMY’s Opdivo and Yervoy feature prominently in the many ongoing checkpoint inhibitor races for as yet unapproved indications, and the Checkmate-026 result does nothing to impair these prospects

Where we’re BULLISH: Biopharma companies with undervalued pipelines (e.g. BMY, GILD, SHPG, SNY, VRTX); Biopharma companies with pending major product approvals (e.g. ACAD, ADMA, ALIOF, BIIB, CHMA, CLVS, CPRX, CTIC, GILD, ICPT, JAZZ, LLY, LPCN, MRK, NVO, OCUL, PTCT, SRPT, TEVA, ZSPH); SNY on sales potential for Praluent (alirocumab); CNC, MOH and WCG on bullish prospects for Medicaid HMOs; and, DVA and FMS for the likely gross margin effects of generic forms of Epogen

Where we’re BEARISH: PBMs facing loss of generic dispensing margin as the AWP pricing benchmark is replaced (e.g. ESRX); Drug Retail as dispensing margins are pressured by narrowing retail networks and replacement of AWP (e.g. WBA, CVS); ABBV on Humira US pricing risks; ENDP on risks to branded Rx price premia; Research Tools & Services companies as growth expectations and valuations are too high in an environment of falling biopharma R&D spend (e.g. CRL, Q, ICLR); and, suppliers of capital equipment to hospitals on the likelihood hospitals over-invested in capital equipment before the roll-out of the Affordable Care Act (e.g. ISRG, EKTAY, HAE)

A narrow look at recent news

BMY’s Opdivo (nivolumab) and MRK’s Keytruda (pembrolizumab) both are PD-1[1] based checkpoint inhibitors, and both currently are approved for the 2nd-line treatment of non-small cell lung cancer (NSCLC) patients whose tumors have progressed on or after platinum-based chemotherapy. Keytruda’s label is further limited to patients whose tumors have PDL1[2] expression levels of 50% or greater; Opdivo’s NSCLC label is not (yet) restricted by PDL1 expression

Both agents are being evaluated in trials intended to strengthen their respective holds on the NSCLC indication, and to increase the numbers of NSCLC patients falling within the bounds of their respective labels. Keytruda recently showed greater progression free and overall survival (PFS, OS) as compared to standard chemotherapy in previously untreated NSCLC patients (i.e. as a 1st-line agent) whose tumors showed PDL1 expression levels of 50% or higher (Keynote-024). Opdivo recently failed a similar 1st-line comparison v. standard chemotherapy (against the PFS endpoint; OS results are not yet known) in previously untreated NSCLC patients whose PDL1 expression levels were just 5% or greater (Checkmate-026)

The obvious difference in the Keynote-024 and Checkmate-026 populations are the PDL1 expression levels required to qualify (50% or greater for Keynote-024 v. 5% or greater for Checkmate-026). This raises the question of whether Opdivo’s failure in Checkmate-026 is attributable to having recruited patients with much lower average PDL1 expression levels – as seems likely[3]

Despite having been approved for 2nd-line NSCLC regardless of PDL1 expression levels, Opdivo’s earlier phase 3 trials gave mixed signals as to whether PDL1 levels were predictive of tumor response. In Checkmate-017[4], Opdivo was superior to docetaxel regardless of PDL1 level. PDL1 expression of 1 percent or greater was present in 53 percent of evaluable tumor samples; however PDL1 expression did not appear to influence overall survival. Conversely in Checkmate-057[5], Opdivo response rates were higher in patients whose tumors had higher PDL1 expression levels, and on par with the comparator (docetaxel) in patients whose tumors were PDL1 negative. What’s more, an arm of the multi-cohort phase 1b pre-cursor (Checkmate-012) to Checkmate-026 evaluated Opdivo as 1st-line treatment (no comparator) in 52 previously untreated squamous and non-squamous NSCLC patients; objective response rates (ORR) were higher in PDL1+ (28%, 9/32) than in PDL1- (14%, 2/14) patients, and the variation in ORR by PDL1 expression hints at a potentially sensitive underlying relationship between PDL1 levels and Opdivo response (Exhibit 1)

On net, the evidence suggests that NSCLC patients with higher PDL1 expression are more likely to respond to agents with PD-1 based mechanisms (e.g. Opdivo and Keytruda); thus Keytruda’s success in Keynote-024 and Opdivo’s failure in Checkmate-026 is potentially – even likely – due to the fact that Keynote-024 included only patients with PDL1 expression of 50% or greater, as opposed to the 5% or greater threshold in Checkmate-026

A broader look at what happens next in NSCLC

Last week’s Checkmate-026 disclosure was limited to the question of whether Opdivo was superior to standard chemotherapy in all patients across all PDL1 levels. The trial includes an evaluation of efficacy by PDL1 expression level, thus whether Opdivo – like Keytruda – can prove superior to standard chemotherapy in 1st-line PDL1+ patients remains an open question to be answered by the trial’s full, detailed results, which in all likelihood will be disclosed this Fall. And, because Checkmate-026 includes patients with lower PDL1 expression levels than Keynote-024, there’s still an opportunity for an advantageous finding relative to Keynote-024, if Opdivo is superior to standard chemotherapy in patients with moderate PDL1 expression (e.g. 25 – 49%). Also, it remains possible that Opdivo could show an overall survival (OS) advantage despite having failed to show a progression free survival (PFS) advantage, though the odds seem remote that such an advantage will be seen across all PDL1 expression levels

The worst case for Opdivo is that the Checkmate-026 findings are too weak to support a competitive label in 1st-line NSCLC patients with PDL1 expression levels above 50%, in which case Keytruda would – at least temporarily — effectively own this patient segment, which accounts for roughly 25% of NSCLC patients overall

The question remains of how PD-1 monotherapy compares to PD-1 in combination with another agent, with or without regard to PDL1 expression level. BMY and AZN appear to be betting that PD-1 / CTLA4 combination therapy is the right strategy with relatively little regard for PDL1 expression levels; conversely MRK and Roche appear to be betting on PD-1 monotherapy in high PDL1 expression patients, and on PD-1 in combination with standard chemotherapy for patients with weaker PDL1 expression. Of the four major PD-1 competitors, BMY appears to be the best hedged. Despite a large bet on the PD-1 / CTLA4 combination, BMY also has begun evaluating Opdivo + standard chemotherapy as a 1st-line NSCLC option. MRK has a relatively small / early NSCLC trial that includes both Keytruda and Yervoy (BMY’s CTLA4). AZN has no trials evaluating durvalumab in combo with traditional chemotherapy, and Roche has no trials evaluating Tecentriq in combination with a CTLA4

Early evidence is supportive of the PD-1 / CTLA4 combination. In yet another arm of Checkpoint-012, BMY has already shown in phase 1b that Opdivo in combination with Yervoy can generate high response rates (57%) as 1st-line monotherapy in NSCLC patients across PDL1 expression levels[6], and that response rates are higher in patients with higher PDL1 levels[7]. A full phase 3 study (Checkmate-227) is underway in 1,980 1st-line NSCLC patients in four arms: Opdivo v. Opdivo + Yervoy v. Opdivo + standard chemotherapy v. standard chemotherapy; the estimated primary completion date is January 2018. MRK is taking a much narrower (308 patients, 2nd-line) and more preliminary (phase 1 / 2) look at Keytruda + Yervoy v. Keytruda + chemotherapy, 2nd-line, in NSCLC patients with recurrent tumors following surgery and adjuvant therapy (NCT02039674). As such, if adding a CTLA4 based mechanism improves on PD-1 based treatment – as evidence suggests – BMY’s clinical strategy still has a very meaningful chance of keeping Opdivo in a predominant NSCLC role, at least relative to Keytruda

If the PD-1 / CTLA4 combination is the preferred approach, AZN likely would become BMY’s main NSCLC competitor. Like BMY, AZN appears to be betting heavily that the PD-1 / CTLA4 combination (durvalumab / tremelimumab) is the right approach for NSCLC. Of the more than 3,300 NSCLC patients AZN is studying in phase 3 trials, 2,622 are evaluating durvalumab and tremelimumab in combination, and the majority of these (1,892) are in trials comparing the durvalumab / tremelimumab combination to standard chemotherapy in 1st-line patients (see appendices for NSCLC trial summaries by agent). Estimated primary completion dates for these trials range from January 2017 (NCT02453282) to October 2018 (NCT02542293)

Roche’s Tecentriq (a PD-1 inhibitor) also has an aggressive NSCLC program, with 6,671 patients across eight phase 3 trials. Like MRK, Roche is evaluating Tecentriq as a 1st-line single agent in NSCLC, and also like MRK is evaluating Tecentriq in combination with chemotherapy as a 1st-line alternative to chemotherapy alone (see appendices for NSCLC trial summaries by agent)

A broader (preliminary) look at what happens next with checkpoint inhibitors

Exhibit 2 summarizes incidence (blue, left-axis, per 100,000 population) and mortality (orange, also left-axis and per 100,000 population) rates for the most significant cancers in which checkpoint inhibitors (CI’s) are currently being evaluated in clinical trials.[8] The mortality rate per diagnosis is provided by cancer type as a grey line, and cancers are sorted from least to most survivable. Cancers for which one or more CI’s are approved are boxed in red. The share of all CI clinical trials, by cancer type, for indications in which the CI being studied is unapproved, are provided as grey columns at the bottom of the exhibit. NSCLC accounts for roughly 21% of all mortality associated with cancers for which CI’s are either approved, or being evaluated in earnest. Roughly one-quarter of NSCLC patients have PDL1 expression levels in excess of 50%, thus these patients likely account for roughly 5% of aggregate mortality

Of the 559 trials in which a CI is being studied for an indication in which that CI is not approved, the great majority of trials (354, 63%) involve an indication in which no CI has yet been approved (i.e. indications other than melanoma, NSCLC, renal cell carcinoma, urothelial carcinoma, Hodgkin Lymphoma, or squamous cell carcinoma of the head and neck (HNSCC)). In aggregate these CI trials are the ‘land-grabs’, i.e. attempts by various CI’s to be the first (or best) to be approved in an indication no CI has yet claimed (Exhibit 3)

This raises the obvious question of which CI’s are most likely to be first – or at least early – in the as yet unclaimed indications. Exhibit 4 summarizes, for each of the unapproved indications, which CI was the first to start clinical trials, is studying the most patients, is conducting the largest number of trials, and/or has reached the most advanced stage of development. BMY’s Opdivo and Yervoy feature prominently, along with MRK’s Keytruda

Exhibit 5 goes a step further, summarizing the active CI’s relative positions by mortality rates of the tumors in which CI’s are being evaluated, but in which no CI has yet been approved. For each indication, we categorized each CI as being either the ‘Leader’, a ‘Close Follower’, or ‘Late’. ‘Leaders’ are defined as the CI’s which, based on registered clinical trials and assumed normal odds of approval by phase, are most likely to be the first CI approved for a given indication. ‘Close Followers’ are defined as CI’s who are likely to have begun phase 3 trials in an indication before the Leader is approved for that indication. ‘Late’ CI’s are those which are unlikely to reach phase 3 before the Leader is approved for a given indication. Percentage values in the table reflect the percent of cancer-related deaths (mortality) that fall into an indication in which a given agent is the Leader, a Close Follower, or Late. BMY’s Opdivo is the lead CI in indications that account for 42.6% of the mortality associated with tumors being evaluated, and a close follower in tumors that account for 49.0%, ranking 2nd on the basis of combined Leader / Close Follower percentage. BMY’s Yervoy is a leader in indications that account for 52.2% of mortality, ranking 4th. MRK’s Keytruda is a Leader in tumors accounting for 28.6% of mortality, and Close Follower in indications that account for 71.4% of mortality, thus ranking 1st


  1. PD-1: programmed cell death 1
  2. PDL1: programmed cell death ligand
  3. Also relevant is that the 1st-line comparator in Checkmate-024 is much tougher than the comparator Opdivo faced in the 2nd-line trials. Opdivo’s earlier NSCLC trials compared Opdivo to docetaxel in patients whose tumors had progressed on or after platinum-based chemotherapy (i.e. 2nd-line) – and in all candor, despite being standard of care docetaxel sets a low bar (circa 7% response rates) in this patient population. In contrast, Opdivo’s comparator (platinum-based chemotherapy) for 1st-line therapy in the Checkmate-026 trial has much higher (25-35%) typical response rates. In fairness, Keytruda faced the same high 1st-line response rates in Keynote-024
  4. Opdivo v. docetaxel, 2nd-line in 272 patients with squamous NSCLC with disease progression on or after platinum-based chemotherapy
  5. Opdivo v. docetaxel, 2nd-line in 582 non-squamous NSCLC patients with disease progression on or after platinum-based chemotherapy
  6. Inclusion criteria included a requirement of PDL1 expression ≥ 1%
  7. 92% (12/13) in patients with PDL1 expression ≥ 50%
  8. Cancers accounting for 80 percent of CI clinical trials

©2016, SSR, LLC, 1055 Washington Blvd, Stamford, CT 06901. 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 Pfizer Inc (PFE) and to Merck (MKGAY) on both securities-related and non-securities-related topics

Print Friendly, PDF & Email