In the updates with 23 October 2015 and 16 March 2015 cut-off dates, results were also presented for OR, ‘CBR’, DoR; with 112 (57% of the required 198 OS events) deaths at the most recent cut-off (23 October 2015), the OS data are too immature for analysis.
In the ITT population OR (CR/PR) rates were higher in the treatment than the control arm, but in the analysis from the data cutoff date of 23 October 2015:
the ITT investigator-assessed OR rate was 21.0% (95% CI: 16.9-25.7) in the palbociclib plus fulvestrant arm and 8.6% (95% CI: 4.9-13.8) in the placebo plus fulvestrant arm;
the odds ratio was 2.78 (95% CI: 1.56-5.60) in favour of palbociclib plus fulvestrant treatment, with a stratified 1-sided p-value of 0.0001.
The sponsor provided additional analyses for endpoints that were not listed in the study design with this latest data cut-off, including the OR for patients with measurable disease.
Table 16: Study A5481023 Summary of objective response, clinical benefit response and duration of objective response – Investigator assessed as of 23 October 2015, ITT population
In the ITT population, 4 patients (2.9%) achieved a CR in the control arm compared with none (0%) in the treatment arm, while 21% in the treatment arm achieved a PR compared with 6.3% in the control arm. SD was reported in 52.5% (42.5%>24 weeks) of the palbociclib and fulvestrant arm compared with 41.3% (31%>24 weeks).
The objective progression rate (progression without any response) was 17% in the palbociclib and fulvestrant arm compared with 33% in the comparator arm.
Comment: For the first time, the OR is now statistically significantly improved in the palbociclib and fulvestrant arm. However, the OR for those with measurable disease was not a secondary endpoint and involves a subgroup analysis that was not prespecified and therefore should not be included in PI and the sponsor is requested to remove this in providing an updated PI.