During an OncLive Peer Exchange® discussion, moderator Mark A. Socinski, MD, and other lung cancer experts from across the United States discuss what, in this context, the PACIFIC trial results mean for practicing oncologists and their patients.
Mark A. Socinski, MD
After a signiciant drought in progress in treating stage III non—small cell lung cancer (NSCLC), the thoracic oncology community is decidedly excited by the recently published results of the phase III PACIFIC trial (NCT02125461).1 The PACIFIC investigators found that administering the immune checkpoint inhibitor (ICI) durvalumab (Imfinzi) to patients with stage III unresectable NSCLC after concurrent chemoradiotherapy (CCRT) significantly enhanced overall survival (OS) with an acceptable increase in toxicity.1
For more than a decade, CCRT has been the standard of care for fit patients with unresectable locally advanced (LA) NSCLC,2-4 and currently, between 30% and 50% of newly diagnosed cases of LA-NSCLC are unresectable.5 Although multiple clinical trials have associated CCRT with greater clinical benefit than radiotherapy alone or sequential radiotherapy and chemotherapy, prognosis after CCRT has remained dismal.2,3,6 Approximately 10% of patients die within 6 months of treatment,4 OS ranges from 18 to 28 months, and the estimated 5-year OS rate is 15% to 25%.6,7
Efforts to improve outcomes by increasing the radiotherapy dose, administering a targeted agent with CCRT or as consolidation therapy, or replacing second-generation chemotherapy agents with newer ones all failed to improve survival.1,6,8 Surgical resection for stage III disease was also unsuccessful at prolonging survival.2 Further, some attempted approaches significantly increased the risk of morbidity or mortality compared with standard CCRT.2,5,8
During an OncLive Peer Exchange® discussion, moderator Mark A. Socinski, MD, and other lung cancer experts from across the United States discussed what, in this context, the PACIFIC trial results mean for practicing oncologists and their patients. The results “have really changed the standard of care in stage III disease,” Socinski said.
The PACIFIC Trial
In recent years, the FDA has approved several ICIs that target PD-1 or PD-L1 for stage IV NSCLC based on clinical trial results that have shown strong efficacy in some patients.6 The activity of PD-1/PD-L1 inhibitors in metastatic NSCLC, coupled with preclinical studies that suggested CCRT amplifies PD-L1 expression in tumors, generated interest in evaluating these ICIs for unresectable LA disease.1 Preliminary data from the phase III investigation of durvalumab in patients with unresectable stage III NSCLC whose disease had not progressed after CCRT led the FDA to approve the anti—PD-L1 monoclonal antibody for this patient population in early 2018.9
Trial Design and Efficacy
“The [PACIFIC] trial was actually a placebocontrolled trial of durvalumab delivered every 2 weeks in a group of patients who had inoperable stage III disease and had completed standard CCRT,” said Mohammad Jahanzeb, MD, describing the study’s design. Adults (median age, 64 years) were recruited at 235 sites in 26 countries spanning every continent except Antarctica.1 Jahanzeb said two-thirds of the approximately 700 patients were randomly assigned to durvalumab (n = 473) and the rest to placebo (n = 236) within 42 days of their last radiation dose.1 Durvalumab was administered intravenously at a dose of 10 mg/kg until disease progression or study withdrawal for up to 12 months.1 He said PACIFIC used a blinded panel to assess OS and progression-free survival (PFS), which were the trial’s primary endpoints.1
Jahanzeb noted that median PFS was almost 3 times longer in the durvalumab arm than in the placebo arm (17.2 vs 5.6 months, respectively; HR for disease progression or death, 0.51; 95% CI, 0.41-0.63).1 He explained that although median PFS of 5.6 months for the control arm appears short, efficacy endpoints were calculated from the time of randomization.1 “When you take that into account, the control arm is not as bad as we think, and it’s believable,” he said.
Median OS was 28.7 months in the control arm and not reached in the durvalumab arm (HR for death, 0.68; 99.73% CI, 0.470-0.997; 2-sided P = .0025),1 and Socinski described median OS in the control arm as “pretty good for any standard” (Table).1 Median time to death or distant metastasis was also significantly longer in the durvalumab arm than in the placebo arm (28.3 vs 16.2 months, respectively; HR, 0.53; 95% CI, 0.41-0.68), and patients in the durvalumab arm were less likely to develop new brain metastases than patients in the placebo group (6.3% vs 11.8%, respectively).1 The findings also suggest that durvalumab may reduce metastatic spread.1 John V. Heymach, MD, PhD, said the results show that durvalumab adds a very clear benefit, representing a tremendous advance for unresectable LA-NSCLC.
Approximately 30% of patients in the durvalumab arm had a grade 3/4 adverse event compared with 26% of patients in the placebo arm.1 Jahanzeb said that although the rates are higher than those observed in studies of ICI monotherapy for stage IV disease, the similar rates in both arms suggest “it’s probably CCRT related—delayed toxicities that are being documented.”
PACIFIC investigators were concerned about the risk of pneumonitis, a frequent complication of both thoracic radiation and PD-1/ PD-L1 inhibitors. Socinski said that results were reassuring because although grade 1/2 pneumonitis was slightly higher in the durvalumab arm, rates of grade 3/4 pneumonitis were similar in the durvalumab and placebo arms (1.9% vs 1.7%, respectively). Rates of grade 3/4 radiation pneumonitis were also higher in the durvalumab arm than in the placebo group (1.5% vs 0.4%).1
Jarushka Naidoo, MBBCh, was involved in a subset analysis of the patients with pneumonitis in PACIFIC that was presented at the 19th World Conference on Lung Cancer.10 She said that a difficulty in evaluating pneumonitis risk is that “we don’t know what are truly the features that will distinguish radiation pneumonitis from immune-related pneumonitis in the context of recent CCRT.”
Implications for Practice
The PACIFIC trial has changed the standard of care for stage IIIB NSCLC, and the panel examined its implications for practicing oncologists. Because every patient is different, oncologists are likely to face questions that PACIFIC does not answer about how to incorporate durvalumab into treatment after CCRT. The panelists discussed how their institutions are dealing with the new challenges.
Timing of Imaging and Durvalumab Initiation
Leora Horn, MD, MSc, FRCPC, said adopting the PACIFIC regimen into practice has required changes to how she schedules follow-up imaging. She said that her institution has started performing the post-CCRT computed tomography scan to evaluate for progression just a few days before starting durvalumab. “We used to wait a few months,” Horn said, explaining that now she often has to justify to insurers why the patient needs a scan only 7 weeks after the previous one. She said she also has to reassure patients “that we’re not looking for a response, because it’s early.”
The panel also discussed the timing of durvalumab after CCRT and dosing frequency. Socinski mentioned that the protocol of PACIFIC was amended from randomizing patients to durvalumab within 2 weeks of completing CCRT to within 4 weeks because of accrual issues. “Some data from that analysis suggest that starting [durvalumab] earlier is better,” he said. Horn said she discusses the data and tries to start durvalumab within 2 to 4 weeks but that some patients are still suffering the effects of CCRT and are not ready.
Calling the 2-week dosing schedule of durvalumab inconvenient compared with other ICIs, Jahanzeb asked whether anyone would be comfortable administering it every 4 weeks. Socinski said his patients have not complained much about the dosing frequency, but Naidoo said she had received pushback from some patients. Her institution is participating in a clinical trial of nivolumab (Opdivo) versus nivolumab/ipilimumab (Yervoy) in which nivolumab is administered every 4 weeks, and Naidoo said “the schedule has been the reason why some people have preferred that” to starting durvalumab.
Adjuvant and Neoadjuvant Durvalumab
The panel debated whether they would choose surgery or durvalumab for a patient whose disease might be operable after CCRT. Heather A. Wakelee, MD, said her institution believes in identifying surgical candidates at diagnosis, before starting treatment. “You don’t want to start with the chemoradiation and then say, ‘Oh, maybe we should go to surgery,’” she noted. “For those who aren’t getting surgery, we absolutely are following the PACIFIC regimen,” Wakelee confirmed.
Heymach said that although an earlier phase III trial found surgical resection after CCRT generally did not improve OS in medically fit patients with stage IIIA (N2) disease, an exploratory analysis concluded that lobectomy after CCRT improved OS and PFS compared with chemoradiotherapy alone.3 The failure of resection to improve OS was due to the high number of deaths in patients who underwent pneumonectomy.3 “So our practice at The University of Texas MD Anderson Cancer Center is if somebody is potentially a surgical candidate, we try to integrate that with other treatments,” he said. Although the PACIFIC trial did not evaluate surgery plus durvalumab, Heymach said practicing clinicians do not always treat patients per study protocol. “If you’ve got somebody who had low-volume mediastinal disease, who responded well to CCRT, and who was a surgical candidate, you do have to make a choice after that surgery…I still lean toward giving them the potential benefits of the adjuvant durvalumab,” he said. Socinski expressed concern about the lack of phase III data establishing the safety of adjuvant ICI use. Wakelee said 4 similarly designed clinical trials are currently assessing adjuvant therapy with PD-1/PD-L1 inhibitors, including nivolumab, pembrolizumab (Keytruda), atezolizumab (Tecentriq), and durvalumab, but only in patients identified as surgical candidates prior to CCRT. “But we don’t really know yet about giving these drugs after surgery—we just don’t have those data back,” she said.
Wakelee said another question that must be answered is how to proceed after administering an ICI. “So many neoadjuvant data are now coming out,” she said, mentioning the phase II NEOSTAR/INDUCTION (NCT03158129) trial that assessed induction therapy consisting of nivolumab with or without ipilimumab in resectable NSCLC.11 The major pathologic response rate was 26% overall.11 “Do we do chemo with immune therapy? Do we do immune therapy alone? Do we bring radiation in before the surgery or more often after the surgery if it’s a known N2?” Wakelee asked. She expressed hope that ongoing trials of neoadjuvant ICI use would answer these questions.
The PACIFIC trial evaluated durvalumab in all-comers regardless of PD-L1 expression or EGFR mutational status.1 “In PACIFIC, I think only 6% of the population had EGFR mutations, [which was] probably not representative of the population,” Naidoo said. The EGFR-positive population was too small for the investigators to draw conclusions about an OS benefit. The study was not designed to analyze outcomes by PD-L1 status, but a post hoc subgroup analysis associated durvalumab with similar improvement in PFS regardless of PD-L1 expression.8 The only difference in OS was observed in patients with PD-L1 expression levels <1%, but the confidence interval was too wide to draw meaningful conclusions.1 “We would certainly offer durvalumab to all our patients, including the stage III patients,” Naidoo said.
Regarding ICI response and PD-L1 and EGFR status, Heymach said, “We can’t extrapolate from the metastatic setting where we’re using [ICI] monotherapy.” He added that data suggest radiation and chemotherapy upregulate PD-L1 expression and affect the tumor microenvironment. “I think it wouldn’t be appropriate to exclude EGFR mutants or PD-L1—negative patients given that it’s a different disease setting,” he concluded.
Wakelee agreed and said she was surprised the European Medicines Agency had restricted approval of durvalumab after CCRT to patients with stage IIII NSCLC who express PD-L1, which she called an imperfect biomarker. “There are absolutely patients without PD-L1 expression who are benefiting in metastatic disease with the checkpoint inhibitors, so it would make sense that we would be seeing the same thing in earlier stages of the disease,” she said.
Wakelee emphasized that driver mutations trump PD-L1 expression levels in patients with lung adenocarcinoma, however, when selecting treatment. She said, “If you have a patient, especially a nonsmoking patient with lung cancer and they have high PD-L1, the reflex should not be just starting them on immune therapy.” She said clinicians should at the very least obtain EGFR and ALK test results. “You don’t want to make that treatment recommendation until you really have the full story,” Wakelee said.
Editor’s Note: This panel discussion took place prior to the FDA approval of pembrolizumab (Keytruda) for stage III NSCLC.