Long-Term Efficacy and Safety of Brigatinib in Crizotinib-Refractory ALK+ NSCLC: Final Results of the Phase 1/2 and Randomized Phase 2 (ALTA) Trials

Introduction We report brigatinib long-term efficacy and safety from phase 1/2 and phase 2 (ALTA) trials in ALK–rearrangement positive (ALK+) NSCLC. Methods The phase 1/2 study evaluated brigatinib 30 to 300 mg/d in patients with advanced malignancies. ALTA randomized patients with crizotinib-refractory ALK+ NSCLC to brigatinib 90 mg once daily (arm A) or 180 mg once daily (7-d lead-in at 90 mg; arm B). Results In the phase 1/2 study, 79 of 137 brigatinib-treated patients had ALK+ NSCLC; 71 were crizotinib pretreated. ALTA randomized 222 patients (n = 112 in arm A; n = 110 in arm B). Median follow-up at phase 1/2 study end (≈5.6 y after last patient enrolled) was 27.7 months; at ALTA study end (≈4.4 y after last patient enrolled), 19.6 months (A) and 28.3 months (B). Among patients with ALK+ NSCLC in the phase 1/2 study, median investigator-assessed progression-free survival (PFS) was 14.5 months (95% confidence interval [CI]: 10.8–21.2); median overall survival was 47.6 months (28.6–not reached). In ALTA, median investigator-assessed PFS was 9.2 months (7.4–11.1) in arm A and 15.6 months (11.1–18.5) in arm B; median independent review committee (IRC)-assessed PFS was 9.9 (7.4–12.8) and 16.7 (11.6–21.4) months, respectively; median overall survival was 25.9 (18.2–45.8) and 40.6 (32.5–not reached) months, respectively. Median intracranial PFS for patients with any brain metastases was 12.8 (9.2–18.4) months in arm A and 18.4 (12.6–23.9) months in arm B. No new safety signals were identified versus previous analyses. Conclusions Brigatinib exhibited sustained long-term activity and PFS with manageable safety in patients with crizotinib-refractory ALK+ NSCLC.

Conclusions: Brigatinib exhibited sustained long-term activity and PFS with manageable safety in patients with crizotinib-refractory ALKþ NSCLC.

Introduction
ALK gene rearrangements are detectable in approximately 3% to 5% of patients with NSCLC. [1][2][3] Treatment with ALK inhibitors is the preferred initial systemic approach for ALK rearrangement-positive (ALKþ) metastatic NSCLC. 4 Crizotinib was the first ALK inhibitor approved by the U.S. Food and Drug Administration (FDA) for the treatment of patients with previously untreated metastatic ALKþ NSCLC. Although crizotinib provides improved efficacy and tolerability compared with chemotherapy, most patients experience disease progression on crizotinib within a year. 5,6 The central nervous system (CNS) is often the first site of disease progression on crizotinib, reflecting inadequate drug penetration into the brain. [7][8][9] Other mechanisms of resistance to crizotinib include the acquisition of secondary mutations in ALK that interfere with crizotinib binding, amplification of the ALK fusion gene, and upregulation of bypass signaling pathways. 10 Several next-generation ALK inhibitors, including alectinib, ceritinib, brigatinib, and lorlatinib, with activity against mechanisms of resistance to crizotinib, have since been developed and approved for use in ALK inhibitor-naive and -resistant NSCLC. Brigatinib first gained approval in 2017 for use in patients with ALKþ NSCLC with disease progression on or intolerance to crizotinib. In 2020, brigatinib was granted full FDA approval for treatment of ALKþ NSCLC on the basis of efficacy and safety results from ALTA-1L, a global randomized phase 3 study comparing brigatinib with crizotinib in patients with tyrosine kinase inhibitor (TKI)-naive ALKþ NSCLC. 11 Brigatinib is a next-generation ALK TKI designed to have potent and broad activity against ALK-positive rearrangements and a range of ALK resistance mutations. [12][13][14] The recommended dose of brigatinib (180 mg once daily with 7-d lead-in at 90 mg once daily) was established in a multinational phase 1/2 study 15 and confirmed in the phase 2 ALTA (ALK in Lung Cancer Trial of AP26113) trial in crizotinib-refractory patients with ALKþ NSCLC. 16,17 Results of interim analyses of each study were previously reported, [15][16][17] revealing high overall and intracranial objective response rates (ORRs) and durable responses with an acceptable safety profile.
Here, we report long-term efficacy and safety results from the final analyses of the phase 1/2 and phase 2 (ALTA) trials of brigatinib, completed more than 5 years after the last patient enrolled in the phase 1/2 study and more than 4 years after the last patient enrolled in the ALTA trial.

Study Design and Patients
Phase 1/2 Study. The phase 1/2 single-arm, open-label trial (ClinicalTrials.gov identifier: NCT01449461) was conducted in the USA and Spain. The methods, the complete protocol, and eligibility criteria have been published previously. 15 The dose-escalation phase (phase 1) enrolled patients with histologically confirmed advanced malignancies other than leukemia. The expansion phase (phase 2) enrolled patients with ALKþ or EGFR T790M-positive NSCLC or other cancers with ALK or ROS1 mutations. Herein, we report long-term outcomes for all patients with ALKþ NSCLC treated with brigatinib in any part of the study. In the doseescalation stage, patients received oral brigatinib at total daily doses of 30 to 300 mg; in the expansion stage, three once-daily oral dosing regimens were assessed: 90 mg once daily, 180 mg once daily, and 180 mg with 7day lead-in at 90 mg. Results revealed that treatment with brigatinib 180 mg once daily with a 7-day lead-in at 90 mg provided increased benefit, while reducing the incidence of early onset pneumonitis and other pulmonary adverse events (AEs) that had been reported in a subset of patients in the dose-escalation and early expansion phases of the phase 1/2 study. 15 ALTA. The phase 2 ALTA trial (ClinicalTrials.gov identifier: NCT02094573) was an open-label, randomized, multicenter, international study. Methods and the complete study protocol and eligibility criteria have been published. 16 Eligible patients (18 y of age) had locally advanced or metastatic ALKþ NSCLC that had progressed while receiving crizotinib; at least one measurable lesion per Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 18 ; and Eastern Cooperative Oncology Group performance status of 2 or less. Patients were stratified by baseline brain metastases status (yes or no) and best previous response to crizotinib (investigatorassessed complete response [CR] or partial response [PR] versus other or unknown response); they were randomized 1:1 to brigatinib 90 mg once daily (arm A) or to 180 mg once daily with a 7-day lead-in at 90 mg (arm B).
In both trials, patients could continue brigatinib until they experienced disease progression or intolerable toxicity. Treatment could be continued after progression at the investigator's discretion if there was evidence of clinical benefit. In ALTA, patients in arm A could transition to brigatinib 180 mg once daily after progression at 90 mg once daily.
Each trial was conducted in compliance with the ethical principles of the Declaration of Helsinki, the International Council for Harmonisation guideline for Good Clinical Practice, and all applicable local regulations. All patients provided written informed consent. All protocols were approved by local institutional review boards or ethics committees at each site.

Assessments
In both studies, disease was assessed according to RECIST version 1.1 18 at baseline and every 8 weeks during treatment (every 12 weeks after cycle 15 in ALTA) and at the end of treatment. In the phase 1/2 study, disease was assessed by the investigators; in ALTA, disease was assessed by the investigators and an independent review committee (IRC). All PRs and CRs were required to be confirmed at least 4 weeks after the initial response. All patients were followed for survival every 3 months for up to 2 years after the initial dose of brigatinib (phase 1/2) or for 2 years after the last patient was enrolled (ALTA). AEs, including laboratory abnormalities, were categorized using the National Cancer Institute Common Terminology Criteria for Adverse Events, version 4.0.
Outcomes Phase 1/2 Study. The investigator-assessed ORR per RECIST version 1.1 18 was the primary outcome for four of the five cohorts of the phase 1/2 expansion phase; the CNS response rate per RECIST version 1.1 was the primary outcome for the cohort of patients with ALKþ NSCLC with active, measurable, intracranial CNS metastases at baseline. "Active" was defined as brain metastases without previous radiotherapy or with investigator-assessed progression after previous radiotherapy. "Measurable" was defined as CNS lesions of 10 mm or more. Secondary outcomes for all cohorts included progression-free survival (PFS), time to progression, overall survival, and safety and tolerability.
ALTA. The primary end point of ALTA was the confirmed ORR, as assessed by the investigator, per RECIST version 1.1. 18 Secondary end points included confirmed ORR, as assessed by the central IRC, per RECIST version 1.1; CNS response (in patients with active brain metastases, intracranial ORR was assessed by the investigator and confirmed by IRC per RECIST version 1.1); time to response; duration of response; disease control rate (the percentage of patients with best response of CR, PR, or stable disease, per RECIST version 1.1); PFS; overall survival; and safety and tolerability.

Statistical Analysis
For the phase 1/2 study, data from all patients with ALKþ NSCLC who received brigatinib in any part of the study were pooled and analyzed for efficacy and safety. For ALTA, efficacy was analyzed in the intention-to-treat (ITT) population (all randomized patients) and safety was evaluated in the safety population (all patients who received 1 dose of brigatinib). For both studies, the exact binomial method was used to calculate confidence intervals (CIs); 97.5% CIs were estimated for the confirmed ORR in ALTA (primary end point) and 95% CIs were used for the other outcomes. Median values and two-sided 95% CIs for time-to-event (duration of response, PFS, and overall survival) analyses were calculated using Kaplan-Meier (KM) methods. Statistical analyses were performed using SAS software (version 9.4, SAS Institute, Inc., Cary, NC).

Patients
Phase 1/2 Study. Between September 20, 2011, and July 8, 2014, a total of 137 patients were enrolled in the phase 1/2 study and received brigatinib at doses ranging from 30 mg to 300 mg daily; 79 patients had ALKþ NSCLC. Of the patients with ALKþ NSCLC, 90% (71 of 79) had previously received crizotinib. Among these 79 patients, the most common brigatinib dosing regimens were 180 mg once daily with 7-day lead-in at 90 mg (n ¼ 28), 180 mg daily (90 mg twice daily or 180 mg once daily; n ¼ 25), and 90 mg once daily (n ¼ 14). The last patient's final visit on the study was on February 18, 2020, approximately 5.6 years after the last patient was enrolled, with a median follow-up of 27.7 months (range: 0.2-88.3). Median duration of brigatinib exposure in the 79 patients with ALKþ NSCLC was 20.0 months (range: 0.03-87.2). There were 10 patients who had no disease progression and were still receiving brigatinib at study end (Fig. 1A). Demographic and clinical characteristics at baseline have been published for both studies. 15,16 Efficacy: Phase 1/2 Study Response Characteristics. Among the 79 patients with ALKþ NSCLC in the phase 1/2 study, the confirmed ORR per investigator assessment was 67% (95% CI: 56-77), with median KM-estimated duration of response of 14.9 months (95% CI: 9.9-29.5) ( Table 1). In the 28 patients with ALKþ NSCLC who received the recommended brigatinib dosing regimen (180 mg once daily with 7d lead-in at 90 mg), the confirmed ORR was 79% (95% CI: 59-92), with median duration of response of 14.8 months (95% CI: 7.9-33.3). Response rates and characteristics were similar for patients with ALKþ NSCLC previously treated with crizotinib (   Intracranial response and PFS data were not collected consistently in the phase 1/2 study and therefore could not be analyzed. Overall Survival. For the 79 patients with ALKþ NSCLC, KM-estimated median overall survival was 47.6 months (95% CI: 28.6-not reached [NR]) and probability of survival at 5 years was 42% (95% CI: 30-55; Fig. 2B and Table 1). In the 71 patients with crizotinib-pretreated ALKþ NSCLC, median overall survival was 30.1 months (95% CI: 21.4-55.0), and 5-year overall survival probability was 35% (95% CI: 22-49). All eight patients with crizotinib-naive ALKþ NSCLC were alive 2 years after the first dose (protocol-specified follow-up period for overall survival).
Exploratory subgroup analyses were performed for confirmed ORR (Fig. 3C), PFS (Supplementary Table 1), and overall survival (Supplementary Table 2) by race (Asian and non-Asian), previous chemotherapy, brain metastases at baseline, and best response with previous crizotinib therapy. There were no notable differences in any of these efficacy parameters between subgroups or when compared with the overall ITT population.

ORR (95 %CI)
Overall Race Asian Non-Asian    [15][16][17] In the phase 1/2 study, 15 of the 79 patients with ALKþ NSCLC died within 30 days of the last dose of brigatinib; two deaths were found to be possibly related to brigatinib (unexpected death on day 568 in a patient receiving 90 mg once daily and sepsis on day 541 in a patient allocated to 180 mg once daily with 7-day lead-in at 90 mg). In ALTA, 36 patients (22 in arm A and 14 in arm B) died within 30 days of the last dose of brigatinib; one death was found to be possibly related to brigatinib treatment (sudden death on day 3 in a patient in arm B).

Discussion
In the final analysis of the phase 1/2 study, brigatinib was found to have sustained long-term activity and PFS in patients with ALKþ NSCLC at a median follow-up of 27.7 months (range: 0.2-88.3) and more than 5 years after the last patient was enrolled. In an earlier report of results from the phase 1/2 study, brigatinib had encouraging CNS activity, with favorable intracranial objective responses and PFS at total daily doses of 90 mg or greater.
The sustained long-term activity of brigatinib in patients with crizotinib-refractory ALKþ NSCLC was confirmed in the final analysis of ALTA at a median followup of 19.6 months (range: 0.1-62.8) in arm A and 28.3 months (range: 0.1-66.8) in arm B, and more than 4 years after the last patient was enrolled. The approved dosing regimen (180 mg once daily with 7-d lead-in at 90 mg; arm B) was associated with numerically higher ORR, PFS, and overall survival than the 90 mg daily dose (arm A).
Brigatinib also exhibited sustained intracranial activity in patients with baseline brain metastases. It seems that patients with brain metastases at baseline had better median overall survival than patients without brain metastases. Nevertheless, PFS rates of these two subgroups do not reveal the same trend. If poststudy treatments are not considered and if brain metastasis is considered as the primary form of ALK TKI failure, these results may not seem as intriguing. One potential explanation is that patients with brain metastases at baseline may seem to have better median overall survival because they were treated with brigatinib despite having confirmed brain metastasis, whereas patients without brain metastases at baseline would have discontinued brigatinib on intracranial disease progression. It is possible that without brigatinib protection, death may occur sooner after intracranial progression.
Brigatinib seems to compare favorably with other TKIs in the second-line setting. In patients with crizotinib-pretreated ALKþ NSCLC, alectinib has an IRCassessed ORR of 51%, median duration of response of 14.9 months, median PFS of 8.3 months, 19 and median overall survival of 29.1 months. 20 Alectinib was associated with an intracranial ORR (by IRC) of 64%, with median duration of intracranial response of 10.8 months, in patients with measurable brain metastases at baseline (by RECIST version 1.1). 21,22 Similarly, ceritinib has an ORR of 39% to 43% (by investigator assessment), median duration of response of 6.9 to 9.7 months, median PFS (by investigator assessment) of 5.7 to 6.7 months, and median overall survival of 14.9 months in patients with ALKþ NSCLC previously treated with chemotherapy and crizotinib 6,23 ; among patients with measurable brain metastases, the intracranial ORR was 35%, with median duration of intracranial response of 6.9 months. 6 Lorlatinib has numerically higher overall (ORR: 73%) and intracranial (70%) response rates in crizotinibpretreated patients, although median PFS (11.1 mo) 24 seems to be shorter than that observed with brigatinib (16.7 mo) and mature overall survival data are not yet available for this setting.
Crizotinib was the first ALK inhibitor to obtain FDA approval for use in patients with treatment-naive ALKþ NSCLC. 25,26 Second-and third-generation ALK TKIs (alectinib, brigatinib, ceritinib, and lorlatinib) have efficacy in the treatment of patients with ALK TKI-naive ALKþ NSCLC and have replaced crizotinib as recommended first-line treatments for patients with ALKþ NSCLC. 11,[27][28][29][30][31] Optimal sequencing of next-generation TKIs in TKI-refractory ALKþ NSCLC has not been established. The phase 2 J-ALTA trial assessed the efficacy of brigatinib in 47 Japanese patients with advanced ALKþ NSCLC refractory to alectinib, with or without previous use of crizotinib. 32 Brigatinib had clinically meaningful efficacy, with an ORR (by IRC) of 34%, median duration of response of 11.8 months, and median PFS (by IRC) of 7.3 months. 32 A multinational phase 2 trial (ALTA-2, NCT03535740) has enrolled 104 patients to investigate brigatinib efficacy and safety in patients with ALKþ NSCLC in the post-alectinib or post-ceritinib setting. 33 The safety profile of brigatinib was consistent with previous reports, with no new safety concerns noted. [15][16][17] The most common AEs were gastrointestinal events and elevated blood creatine phosphokinase levels. There were no changes in the incidence of pulmonary AEs with early onset because results were reported in previous publications. [15][16][17] In ALTA, dose reductions were more common at the phase 2 recommended dose of 180 mg once daily after a 7-day lead-in at 90 mg, but these did not seem to compromise efficacy.
In conclusion, brigatinib had sustained long-term activity, PFS, and manageable safety in patients with ALKþ NSCLC. The 180 mg daily dose after 7-day lead-in at 90 mg was associated with numerically longer median PFS and overall survival than the 90-mg daily dose. Final efficacy results of the phase 1/2 and phase 2 (ALTA) trials of brigatinib are similar, if not superior, to those reported for other approved ALK TKIs in the second-line setting. These data and the prospect of prolonged survival in this setting cement the role of next-generation ALK TKIs such as brigatinib in the treatment of patients with advanced ALKþ NSCLC.