Durvalumab as third-line or later treatment for advanced non-small-cell lung cancer (ATLANTIC): an open-label, single-arm, phase 2 study
Garassino MC et al., Lancet Oncol. 2018 Apr;19(4):521-536.
PubMed ID: 29545095
Patients with advanced NSCLC who progressed on two lines of therapy have limited treatment options including erlotinib, gemcitabine and vinorelbine. The advent of immune checkpoint inhibitors has brought about significant strides in improving survival in advanced (EGFR-/ALK-) NSCLC, particularly in tumors with high PD-L1 expression. Durvalumab is an anti-PD-L1 monoclonal antibody that demonstrated antitumor activity in advanced NSCLC in preclinical studies. To further investigate the efficacy and safety of durvalumab as third line or beyond treatment in metastatic NSCLC, researchers conducted the ATLANTIC trial, looking at three patient cohorts based on EGFR/ALK status and tumor PD-L1 expression.
- Patients: histologically or cytologically confirmed locally advanced or metastatic NSCLC (stage III or IV), progressed on at least two lines of therapy, with measurable disease, WHO performance status 0-1, treated or asymptomatic brain or spinal metastases (off corticosteroids), previous treatment with EGFR or ALK inhibitors if positive, at least 25% tumor cells with PD-L1 expression (amendment 1) or EGFR-/ALK- but with 90% tumor PD-L1 expression (amendment 2).
- Cohort 1: EGFR+/ALK+ NSCLC with ≥25% or < 25% tumor PD-L1 expression
- Cohort 2: EGFR-/ALK– NSCLC with ≥ 25% or <25% tumor PD-L1 expression
- Cohort 3: EGFR-/ALK- NSCLC with ≥ 90% tumor PD-L1 expression
- Study design: phase II open-label, single-arm trial
- Durvalumab 10mg/kg IV q2w
- Primary: ORR
- Secondary: OS, PFS, DOR, DCR, TTR
- 444 patients treated, cohort 1 (n=111), cohort 2 (n=265), cohort 3 (n=68)
- Mutation status
- Cohort 1: 87% (91) EGFR+, 14% (15) ALK+, 1% (1) EGFR+ and ALK+
- Cohort 2: 0% EGFR+, 0% ALK+, 0% EGFR+ and ALK+
- Cohort 3: 0% EGFR+, 0% ALK+, 0% EGFR+ and ALK+
- More Asian patients enrolled in cohort 1
- Cohort 3 had the longest median duration of exposure to durvalumab: 24.1 wks. vs. 12 wks. (cohort 1) vs. 16.1 wks. (cohort 2)
- 23% received subsequent therapy, mostly with erlotinib (5%) while 14% received subsequent RT
Table 1: Efficacy Outcomes
Table 2: Adverse events and treatment summary
- Grade 3 adverse events occurred in ≤1% in cohorts 1 and 2 except for grade 3 increased AST and diarrhea in cohort 3 (3% each).
- Grade 4 lymphocyte count reduction and tumor hemorrhage occurred in 1(1%) each in cohort 1 while grade 4 lymphopenia and increased gamma-glutamyltransferase occurred in 1(1%) each in cohort 2.
- Grade 1-2 and 3 adrenal insufficiency occurred in <1% each in cohort 2.
- In cohort 1, one patient (1%) developed grade 3 spinal compression while another patient (1%) had grade 3 pericardial effusion.
- The most common serious adverse events were pneumonitis, fatigue and infusion-related reaction (1% each).
- Treatment discontinuation was most commonly due to grade 2-3 pneumonitis (n=3), elevated hepatic enzymes (n=2), anemia (n=1), hypovolemic shock (n=1), nephritis (n=1), infusion-related reaction (n=1), and diarrhea (n=1).
- No grade 4-5 immune-related adverse events occurred during the study.
- One patient who died from pneumonitis was deemed to be possibly drug-related.
- Trough levels of durvalumab were similar in all three cohorts at all time points.
- Immunogenicity occurred in 16 (4%) overall
- In the ATLANTIC study, a significant proportion of patients (12-31%) with high PD-L1 expression achieved objective responses as early as 1.8 months with durvalumab in heavily pretreated advanced NSCLC. Although the EGFR-/ALK- NSCLC with ≥ 90% tumor PD-L1 expression group (cohort 3) achieved the highest tumor response, these patients were less heavily pretreated compared to the other 2 cohorts.
- DOR was comparable for all cohorts, regardless of tumor PD-L1 expression level. However, DCR was higher in the EGFR-/ALK- cohort and those with higher PD-L1 tumor expression (≥25% and ≥90%). Patients with higher tumor PD-L1 expression had greater tumor responses, including best change in tumor size from baseline. Similarly, this group had longer OS, as well as EGFR-/ALK- NSCLC patients compared to those with lower tumor PD-L1 expression or EGRF+/ALK+ NSCLC.
- Durvalumab was generally well tolerated, with a toxicity profile consistent with other anti-PD-1/PD-L1 inhibitors. Most immune and non-immune mediated adverse events were low grade. Although cohort 3 was associated with the best clinical outcomes, this group also had the highest rates of high grade and severe adverse events as well as immune-mediated adverse events which could possibly be related to longer duration of treatment compared to the other cohorts (24.1 wks. vs. 12 wks. and 16.1 wks. respectively).
- CheckMate 063 trial investigated the efficacy of nivolumab as third line and later treatment in advanced, refractory NSCLC (Lancet Oncol 2015;16: 257–65). At least 14.5% achieved an objective response and 24% with at least 5% tumor PD-L1 expression responded to treatment. Median TTR was 3.3 mos., with median PFS of 1.9 mos. and median OS of 8.2 mos. ORR in this study was similar to that achieved in cohort 2, although occurring later and with shorter OS compared to findings in the ATLANTIC study.
- Pembrolizumab also demonstrated good clinical activity in refractory advanced NSCLC. In the KEYNOTE 001 trial, the proportion who achieved an objective response was 18%, with 3 months median PFS and 9.3 months median OS in previously treated patients (N Engl J Med 2015; 372: 2018–28). Again, ORR rates were similar to cohort 2 and PFS and OS rates appeared to be comparable with the <25% tumor PD-L1 expression cohorts in the ATLANTIC study.
- ATLANTIC study limitations include a lack of comparator arm, relatively small sample size (especially cohort 3), inconsistency in follow-up duration between cohorts, short follow-up period and lack of formal statistical comparisons between cohorts. Its strengths are the use of durvalumab in the third line setting and beyond where treatment options are limited and focus on EGFR/ALK mutation status and tumor PD-L1 expression.
- In summary, treatment with durvalumab resulted in good clinical activity and durable tumor responses in heavily pretreated NSCLC, with enhanced activity in tumors having higher PD-L1 expression, regardless of EGFR/ALK mutation status. The possible benefit in EGFR+/ALK+ tumors with high PD-L1 expression warrants further investigation.
More Lung Cancer Updates
Venetoclax–Rituximab in Relapsed or Refractory Chronic Lymphocytic Leukemia
Seymour JF et al., N Engl J Med. 2018 Mar 22;378(12):1107-1120.
PubMed ID: 29562156
Venetoclax is an oral, highly selective BCL2 inhibitor with good antitumor activity in heavily pretreated chronic lymphocytic leukemia (CLL). Studies on the combination of venetoclax with rituximab in CLL revealed clinically significant tumor responses and manageable toxicity. In the phase III MURANO trial, researchers explored the efficacy and safety of this combination versus bendamustine with rituximab in relapsed or refractory CLL. Results from interim analysis are summarized below.
- Patients: relapsed/refractory CLL, received 1-3 previous therapies, Eastern Cooperative Oncology Group (ECOG) performance status score 0-1, previous bendamustine allowed if duration of response (DOR) after treatment was at least 24 months.
- Study design: phase III open-label RCT
- Experimental: Venetoclax (20 to 400 mg/day ramp-up) up to 2 yrs. + rituximab 375 mg/m2 cycle 1 then 500 mg/m2 subsequent cycles every 28 days
- Control: Bendamustine 70 mg/m2 every 28 days + rituximab 375 mg/m2 cycle 1 then 500 mg/m2 subsequent cycles every 28 days
- Primary: progression-free survival (PFS) - investigator-assessed
- Secondary: PFS – independent review committee (IRC), overall response rate (ORR), complete response rate (CR), overall survival (OS), rates of clearance of minimal residual disease [MRD] (to below the threshold of 1 tumor cell per 104 white cells), DOR, event-free survival (EFS), time to next treatment for CLL
- At a median follow-up of 28.3 mos., venetoclax-rituximab significantly prolonged PFS compared to bendamustine-rituximab (not reached vs. 17.2 mos. (HR 0.17; p=<0.001). 2-yr. PFS more than doubled with venetoclax-rituximab (84.9 vs. 36.3 mos.).
- The benefit of venetoclax-rituximab was homogeneous across all prespecified subgroups. 2-yr. PFS (investigator-assessed) was consistently higher with venetoclax-rituximab vs. bendamustine-rituximab in patients with chromosome 17p deletion (81.5 vs. 27.8%, HR 0.13) and without chromosome 17p deletion (85.9 vs. 41%, HR 0.19).
- No significant difference was observed in the rate of CR or CR with incomplete hematologic recovery for both groups. IRC-assessed CR or CR with incomplete hematologic recovery was 8.2 vs. 3.6%. ORR by IRC (92.3 vs.72.3%) and investigator assessment were similar (93.3 vs. 67.7%).
- Venetoclax-rituximab provided a higher rate of clearance of MRD in peripheral blood samples at 9 mos. (62.4 vs. 13.3%) over bendamustine-rituximab that was sustained over time. A similar pattern was observed in the rate of clearance of MRD in bone marrow aspirate in favor of venetoclax-rituximab (27.3 vs. 1.5%).
- Survival rates were higher in the venetoclax-rituximab group. OS at 24 mos. was 91.9 vs. 86.6%, respectively (HR 0.48). 2-yr. EFS was also better in the venetoclax-rituximab group (84.9 vs. 34.8%, HR 0.17). Venetoclax-rituximab extended time to next treatment for CLL. At 2 years, 90% did not receive subsequent therapy compared with 52.1% with bendamustine-rituximab (HR 0.19). At least 1.5% in the venetoclax-rituximab group received subsequent targeted therapy vs. 20.5% in the bendamustine-rituximab group.
- Neutropenia was the most common adverse event in both groups (any grade: 60.8 vs. 44.1%, grade 3-4: 57.7 vs. 38.8%), leading to dose interruptions despite the use of G-CSF (47.9 vs. 43.1%). Although these rates were higher with venetoclax-rituximab, grade 3-4 febrile neutropenia and infections/infestations occurred with less frequency compared to the bendamustine-rituximab group. Grade 3-4 tumor lysis syndrome occurred in 6 (3.1%) vs. 2 (1.1%), respectively. Serious adverse events (46.4 vs.. 43.1%), Richter’s transformation (n=6 vs. 5) and adverse events causing death (5.2 vs. 5.9%) occurred with similar frequency in both groups.
- Vnentoclax-rituximab demonstrated superior clinical outcomes in terms of PFS, OS, EFS, tumor response, rate of clearance of MRD and time to next treatment for CLL over bendamustine-rituximab in relapsed and refractory CLL. Myelosuppression occurred despite supportive therapy but rates of infections/infestations and febrile neutropenia were less commonly observed compared to bendamustine-rituximab. A more proactive approach towards supportive management could potentially mitigate hematologic toxicity with this treatment combination.
- In summary, the MURANO study provides evidence for the role of venetolcax-rituximab as a valuable treatment option in relapsed or refractory CLL.
More Hematologic Cancer Updates