Key data read outs at the ESMO congress were dominated by the PD-(L)1 inhibitor class, and the battle for future market share in first- and second-line non-small-cell lung cancer (NSCLC). Novartis – in breast cancer – and Tesaro – in ovarian cancer – also delivered compelling updates.
Merck & Co. – Keytruda (first-line NSCLC)
Merck & Co.'s Keytruda should become the first PD-1 inhibitor to be approved for first-line NSCLC by year-end; specifically in patients whose tumour PD-L1 expression is 50 percent or higher (ViewPoints: Could festive cheer crown 2016 as the year of Keytruda?).
Positive data supporting this indication (from the Keynote-024 study) was top-lined in June and presented in more detail this weekend at ESMO. Median progression-free survival (PFS) was 10.3 months for Keytruda versus 6 months for chemotherapy, with a hazard ratio of 50 percent. Median overall survival data is not yet mature, but at six months 80.2 percent of patients treated with Keytruda were alive versus 72.2 percent of patients who received chemotherapy. Twenty-seven percent of Keytruda patients experienced grade 3-5 side effects, versus 53 percent of chemotherapy patients.
Merck estimates that around 25,000 new NSCLC patients with tumours expressing levels of PD-L1 of at least 50 percent are diagnosed each year, with approximately 22,000 patients in the EU meeting this criteria. A key question is how closely real-world usage of Keytruda will conform to the 50 percent PD-L1 cut-off; feedback from oncologists who were polled by FirstWord in August suggest use could be more liberal.
Bristol-Myers Squibb – Opdivo (first-line NSCLC)
In contrast to Keytruda, Bristol-Myers Squibb's PD-1 inhibitor Opdivo has failed to demonstrate a benefit over chemotherapy in first-line NSCLC patients; top-line data from the CheckMate -026 study revealed this in August.
Versus Merck's more conservative strategy of targeting patients with PD-L1 expression rates of at least 50 percent, Bristol-Myers Squibb enrolled a much broader population with a minimum PD-L1 cut-off rate of 5 percent. Presentation of data on Sunday revealed that Opdivo delayed tumour progression by 4.2 months versus 5.9 months for chemotherapy.
It was hoped that when full data from CheckMate -026 was presented at ESMO, Opdivo would demonstrate comparable efficacy to Keytruda in patients with greater than 50 percent PD-L1 expression levels, but this was not the case. Instead there was no difference in tumour progression or survival rates between patients receiving Opdivo or chemotherapy. This would appear to rule out any slim chance that Bristol-Myers Squibb previously had to secure a compendia listing for Opdivo in higher expressing first-line patients or gain significant off-label use in this indication.
Speaking on separate investor calls on Sunday, management teams from both Bristol-Myers Squibb and Merck could offer little explanation as to why Keytruda had performed so strongly and Opdivo had not even demonstrated a favourable trend towards progression benefit in higher-expressing PD-L1 patients. Could the two antibodies be more different than experts previously thought?
AstraZeneca – durvalumab/tremelimumab (first-line NSCLC)
AstraZeneca did not present key immunotherapy data at ESMO, but comments made by CEO Pascal Soriot on the sidelines caught the attention.
Analysts have speculated for some time that the failure of CheckMate -026 could provide AstraZeneca a greater than expected opportunity in first-line NSCLC; both for its PD-L1 inhibitor durvalumab as a monotherapy and its combination of durvalumab and the CTLA-4 inhibitor tremelimumab. In a recent note to investors, analysts at Jefferies suggested that positive data from AstraZeneca's MYSTIC study could give the company's combination a 12-month head-start on Bristol-Myers Squibb's own PD-1/CTLA-4 combination of Opdivo and Yervoy (which is now the company's primary means of accessing the first-line NSCLC market).
Speaking for the first time since CheckMate -026 was top-lined in August, Soriot said at ESMO "suddenly, this trial news opens quite some opportunities for us, in both monotherapy and combination therapy," reported Reuters. As speculated on by analysts, Soriot also suggested that Opdivo's negative data set and Keytruda's positive data set would allow AstraZeneca to redefine the PD-L1 cut-off levels used in the MYSTIC study before results are un-blinded. In Soriot's words "it gives us a chance to do what Bristol-Myers Squibb was trying to do, but do it in a less risky manner, by expanding the population from the 50 percent PD-L1 cut-off to a bit broader."
Merck – Keytruda plus chemotherapy (first-line NSCLC)
In contrast to the PD-(L)1/CTLA-4 combinations being developed by Bristol-Myers Squibb and AstraZeneca, Merck and Roche are prioritising the development of PD-1/PD-L1 and chemotherapy pairings for the treatment of first-line NSCLC patients.
Data presented to date has been promising, but in small patient numbers. Coinciding with ESMO, however, full results from Merck's Keynote-021 study have been published; these are the first data to be presented for a PD-1/chemotherapy combination trial with an active comparator.
Demonstrating an overall response rate of 55 percent versus 29 percent for chemotherapy monotherapy, the combination of Keytruda plus chemotherapy appears to be "competitive," with the PD-1/CTLA-4 combinations, remarked analysts at Bernstein. They suggest that "these data – if confirmed by improvements in overall survival in ongoing Phase III trials – suggest a future treatment paradigm in which chemo combo could be a preferred approach for PD-L1-negative patients, and might be an option for PDL1 >50 percent patients."
Speaking on an investor call on Sunday, Merck management did not rule out the possibility that as a randomised clinical trial, Keynote-021 may support a compendia listing; if this was to occur it could boost Keytruda sales meaningfully above current consensus, add analysts at Bernstein.
Roche – Tecentriq (second-line NSCLC)
By the end of the month Roche's Tecentriq should become the third PD-(L)1 inhibitor to gain approval in second-line NSCLC. Opdivo has dominated this indication to date, by virtue of a broad label that requires no PD-L1 testing prior to approval, and streaked ahead of Keytruda.
Data presented from Roche's OAK study at ESMO confirmed Tecentriq provides a statistically significant survival benefit in second-line patients irrespective of PD-L1 status (although higher PD-L1 expressers demonstrated a notably superior survival benefit).
Merck has argued that increased PD-L1 testing in first-line patients (once Keytruda secures approval in this indication) will remove the primary impediment to Keytruda second-line use, while the second-line market will also shrink. Furthermore, in low and non-expressing second-line patients, Tecentriq looks poised to offer significant competition to Opdivo; as predicted when data from OAK was top-lined last month (ViewPoints: Roche throws curveball with Tecentriq data).
Novartis – ribociclib (first-line breast cancer)
Beyond the PD-(L)1/NSCLC market, Novartis delivered notable results for its CDK4/6 inhibitor ribociclib in first-line breast cancer.
Ribociclib appears to have taken the lead over Eli Lilly's abemaciclib as the second drug in this class to reach the market; after Pfizer's Ibrance. Given rapid adoption of Ibrance to date, distinction between second- and third-to-market status may prove critical from a commercial perspective.
Based on a comparison of data from Novartis' MONALESSA-2 study and Pfizer's PALOMA trials, analysts at Bernstein described ribociclib and Ibrance as "twins separated at birth." With consensus forecasts for Ibrance in 2020 swelling to around $5.5 billion, Novartis investors will be hoping that ribociclib can also achieve blockbuster status as use of these agents continues to grow.
Tesaro – niraparib (second-line maintenance ovarian cancer)
Expectation for the PARP inhibitor class was high going into ESMO and Tesaro delivered convincing data for niraparib as a maintenance treatment for all patient subgroups included in the Phase III NOVA study.
In patients with germline BRCA mutations, niraparib demonstrated a median PFS of 21 months versus 5.5 month for placebo, a median PFS of 9.3 months versus 3.9 months for placebo in the non-germline BRCA mutation cohort and a median PFS of 12.9 months versus 3.8 months in non-germline BRCA patients whose tumours contain another biomarker called HRD (12.9 months versus 3.8 months for placebo).
Inclusion of HRD-positive patients on niraparib's label would notably expand the commercial opportunity for niraparib, to approximately 50 percent of the ovarian cancer market. However, in demonstrating a statistically significant median PFS benefit of 3.1 months in non-BRCA mutation, HRD-negative patients, study investigator Mansoor Riva Mirza has argued that niraparib should be offered to all ovarian cancer patients. This would render the HRD test developed by Myriad Genetics in conjunction with niraparib unnecessary; unsurprisingly, Myriad has argued that less robust efficacy in HRD-negative patients must be considered in light of the overall toxicity profile in a maintenance indication."
Given that experts were far from convinced approval would be secured in HRD-positive patients on the strength of top-line data from the NOVA study, Tesaro looks to be in a strong position exiting ESMO even if approval in all-comer ovarian cancer patients cannot be secured.
To read more List articles, click here.