Real-world outcomes with durvalumab after chemoradiotherapy in patients with unresectable stage III NSCLC: interim analysis of overall survival from PACIFIC-R

Abstract

Background: Based on the findings of the PACIFIC trial, consolidation durvalumab following platinum-based chemoradiotherapy (CRT) is a global standard of care for patients with unresectable, stage III non-small-cell lung cancer (NSCLC). An earlier analysis from the ongoing PACIFIC-R study (NCT03798535) demonstrated the effectiveness of this regimen in terms of progression-free survival (PFS). Here, we report the first planned overall survival (OS) analysis.

Patients and methods: PACIFIC-R is an observational/non-interventional, retrospective study of patients with unresectable, stage III NSCLC who started durvalumab (10 mg/kg intravenously every 2 weeks) within an AstraZeneca-initiated early access program between September 2017 and December 2018. Primary endpoints are OS and investigator-assessed PFS, estimated using the Kaplan-Meier method.

Results: By 30 November 2021, the full analysis set included 1154 participants from 10 countries (median follow-up in censored patients: 38.7 months). Median OS was not reached, and the 3-year OS rate was 63.2% (95% confidence interval 60.3% to 65.9%). Three-year OS rates were numerically higher among patients with programmed death-ligand 1 (PD-L1) expression on ≥1% versus <1% of tumor cells (TCs; 67.0% versus 54.4%) and patients who received concurrent CRT (cCRT) versus sequential CRT (sCRT) (64.8% versus 57.9%).

Conclusions: PACIFIC-R data continue to provide evidence for the effectiveness of consolidation durvalumab after CRT in a large, diverse, real-world population. Better outcomes were observed among patients with PD-L1 TCs ≥1% and patients who received cCRT. Nevertheless, encouraging outcomes were still observed among patients with TCs <1% and patients who received sCRT, supporting use of consolidation durvalumab in a broad population of patients with unresectable, stage III NSCLC.

Keywords: PD-L1; durvalumab; immunotherapy; locally advanced NSCLC; real-world evidence.

Figure 1

OS and investigator-assessed PFS in the full analysis set. Shown are Kaplan–Meier distributions of (A) OS and (B) investigator-assessed PFS. The tick marks on each trendline represent censored observations, and the dashed lines arising from the x-axes represent 2- and 3-year landmark analyses. OS was defined as the time from the index date to the date of death from any cause, or the last recorded date the patient was known to be alive for censored patients. PFS was defined as the time from the index date to the date of investigator-determined disease progression, or death from any cause (in the absence of progression), whichever occurred first; if no progression or death occurred, patients were censored at the time of the last available tumor assessment. Note: given the real-world nature of PACIFIC-R, progression could be determined by either investigator’s assessment or according to the Response Evaluation Criteria in Solid Tumors version 1.1, depending on local practice. The median duration of follow-up in patients censored at the time of database cut-off was 38.7 months (range 13.6-49.0 months). CI, confidence interval; NE, not estimable; OS, overall survival; PFS, progression-free survival.

Figure 2

OS for subgroups of interest. Shown are Kaplan–Meier distributions of OS according to (A) PD-L1 expression level; (B) prior CRT type; (C) time elapsed between the end of RT and the start of durvalumab; (D) disease stage; and (E) tumor histologic type. The tick marks on each trendline represent censored observations, and the dashed lines arising from the x-axes represent 2- and 3-year landmark analyses. CI, confidence interval; cCRT, concurrent chemoradiotherapy; NE, not estimable; NR, not reached; OS, overall survival; PD-L1, programmed death-ligand 1; RT, radiotherapy; sCRT, sequential chemoradiotherapy; TC, tumor cell.

Figure 3

Investigator-assessed PFS for subgroups of interest. Shown are Kaplan–Meier distributions of investigator-assessed PFS according to (A) PD-L1 expression level; (B) prior CRT type; (C) time elapsed between the end of RT and the start of durvalumab; (D) disease stage; and (E) tumor histologic type. The tick marks on each trendline represent censored observations, and the dashed lines arising from the x-axes represent 2- and 3-year landmark analyses. Note: given the real-world nature of PACIFIC-R, progression could be determined by either investigator’s assessment or according to the Response Evaluation Criteria in Solid Tumors version 1.1, depending on local practice. CI, confidence interval; cCRT, concurrent chemoradiotherapy; NE, not estimable; NR, not reached; PFS, progression-free survival; PD-L1, programmed death-ligand 1; RT, radiotherapy; sCRT, sequential chemoradiotherapy; TC, tumor cell.

Figure 4

TTDM, TDLR, and TFST in the full analysis set. Shown are Kaplan–Meier distributions of (A) TTDM, (B) TDLR, and (C) TFST. The tick marks on each trendline represent censored observations, and the dashed lines arising from the x-axes represent 2- and 3-year landmark analyses. TTDM was defined as the time from the index date to the date of first distant metastasis, as determined by the investigator, or death in the absence of distant metastasis; patients who had not developed a distant metastasis, and who had not died at the time of the analysis, were censored for TTDM at the time of their last available tumor assessment. TDLR was defined as time from the index date until the date of first documentation of local recurrence (i.e. recurrence within the thoracic radiation field), as determined by the investigator; local recurrence was only applicable when there was no distant relapse at the same time or before the local recurrence (in the latter case, it was considered distant relapse and the censoring date was the date of diagnosis in the metastatic setting). Patients who did not experience a local recurrence, or experienced local recurrence and distant metastasis at the same time, or experienced distant metastasis only, were censored; death in the absence of both local recurrence and distant metastasis was considered an event. TFST was defined as time from the index date until the start date of the first subsequent treatment after durvalumab discontinuation or death from any cause, whichever occurred first. CI, confidence interval; NE, not estimable; TDLR, time to death or local recurrence; TFST, time to death or first subsequent treatment after durvalumab; TTDM, time to death or distant metastasis.