Second- and third-generation ALK inhibitors for non-small cell lung cancer

Crizotinib as the first-generation ALK inhibitor has shown significant activity in ALK-mutated non-small cell lung cancer (NSCLC). Second- and third-generation ALK inhibitors are entering clinical applications for ALK+ NSCLC. In addition, a third-generation ALK inhibitor, lorlatinib (PF-06463922), was reported to resensitize NSCLC to crizotinib. This review provided a summary of clinical development of alectinib, ceritinib, brigatinib (AP26113), and lorlatinib.

ALK + NSCLC characteristics ALK-positive NSCLCs are generally seen in non-smokers, occur at a younger age, and are mostly adenocarcinoma in histology [15]. They also seem to have a female gender predisposition [11][12][13]27]. Pathological features include solid morphology and presence of signet ring cells [29,30].
PROFILE 1014 study compared crizotinib vs chemotherapy in 343 patients who had no previous treatment for advanced NSCLC. They were randomized to either receive crizotinib vs pemetrexed plus platinum (cisplatin or carboplatin). Progression-free survival for crizotinib group (n = 172) was 10.9 months and for chemotherapy group (n = 171) was 7.0 months. The ORR was 74 % (95 % CI 67-81) for crizotinib group vs 45 % (95 % CI 37-53) for chemotherapy (P < 0.001). Median OS was not reached in either group at the time of report (hazard ratio for death with crizotinib, 0.82; 95 % CI, 0.54 to 1.26; P = 0.36); the 1-year estimated survival was 84 % with crizotinib vs 79 % with chemotherapy. Crizotinib-associated AEs were vision disorders, diarrhea, nausea, and edema. It was concluded from PROFILE 1014 study that crizotinib was superior to standard first-line pemetrexed-plus-platinum chemotherapy in patients with previously untreated advanced ALKpositive NSCLC. Hence, crizotinib is currently approved for first line in ALK+ NSCLC [14,36].
Crizotinib has also been shown to be highly efficacious in ROS1-positive NSCLC which comprises 1 % of all NSCLC. In a phase 1 study of 50 patients, the ORR was 72 % (95 % CI 58-84) (33 PR and 3 CR). The median PFS was 19.2 months [31]. Among 30 tumors that were tested, 7 ROS1 fusion partners were identified, 2 of these partner genes were novel. However, there was no correlation between the type of ROS1 rearrangement and the clinical response to crizotinib. ROS1 rearrangement molecularly marks a small subgroup of NSCLC for which crizotinib can play an active role in clinical therapy.

Resistance to crizotinib
Majority of patients develop resistance to crizotinib within 1 to 2 years from the initiation of therapy [37]. The resistance to ALK inhibitors can be classified into primary and secondary resistance [38].
Primary resistance is seen when the tumor is deemed refractory to the agent at the beginning of the therapy itself as reported in chronic myeloid leukemia [39]. In the case of ALK+ NSCLC, the primary resistance can be attributed to the different fusion variants of EML4 with ALK or other partner genes [38]. Different sensitivities to crizotinib have been shown to be dependent upon the ALK variant or fusion gene partner [40,41]. Currently, FISH has been the gold standard for detecting ALK mutations in NSCLC.
Secondary resistances are acquired mechanisms after the tumor has been exposed to an ALK inhibitor and can be further classified into two categories: ALK dominant and ALK non-dominant. In the ALK dominant type, there is mutation in the target ALK gene resulting in inability to inhibit the encoded tyrosine kinase. These are termed as ALK dominant as they depend upon ALK tyrosine kinase activity [42]. Most of the mutations are in the form of point mutations and the first ones to be described are C1156Y and L1196M [43]. There have been several other secondary point mutations that have been identified and are the following: G1269A, F1174L, 1151Tins, L1152R, S1206Y, I1171T, G1202, D1203N, and V1180L [41][42][43][44].
The ALK non-dominant resistance involves emergence of bypass tracks such as EGFR mutation, KRAS mutation, amplification of KIT, phosphorylated amplification of ErbB, MET, and activation of IGF-1R in the downstream signaling. It has been shown that in the same ALK resistant tumor, multiple mechanisms of resistances may occur [42,45].
Secondary mutations of the ALK gene result in 29 % of resistant cases, and gene amplification is implicated in 9 % of these cases. The remaining of the cases can be attributed to bypass pathways and other mechanisms that have yet to be defined [46].

CNS metastasis
Crizotinib has poor activity against CNS metastasis in NSCLC as evidenced by low concentrations detected in CNS samples during the course of systemic chemotherapy. The ratio of CNS to serum concentration of crizotinib has been in the range of 0.0006-0.001 as established by individual case reports [47][48][49]. In a retrospective analysis of trials involving crizotinib, 20 % of patients who did not have CNS disease at the beginning had CNS metastasis while on therapy [50]. PF-06463922 (lorlatinib) is a newly developed ALK inhibitor that has been designed for better CNS penetration and is currently in phase I/II trials (NCT01970865) (see below) [15].
In another analysis of 90 patients with brain metastases from ALK-mutated NSCLC, 84 of 90 patients received radiotherapy to the brain, and 86 of 90 received TKI therapy [51]. Significant improvement in this population of poor-prognostic patients was reported. The median OS after development of brain metastases was 49.5 months (95 % CI, 29.0 months to not reached), and median intracranial PFS was 11.9 months (95 % CI, 10.1 to 18.2 months). Four groups of patients were classified in this analysis with distinct outcomes: absence of extracranial metastases, high Karnofsky performance score ≥90, and no prior therapy with TKIs before development of brain metastases had longer survival (P = .003, <.001, and <.001, respectively), whereas isolated brain metastasis or initial treatment with radiation were not (P = .633 and .666, respectively). It was concluded that brain radiotherapy and TKIs to control intracranial disease in ALK+ NSCLC can lead to prolonged survival. Newer TKIs are playing an important role in this population of patients.

Second-generation ALK inhibitors
Ceritinib Within the first year or two after crizotinib treatment is initiated, resistance typically arises. As mentioned above, mechanisms commonly include secondary mutations within the ALK tyrosine kinase domain and activation of alternative signaling pathways. More potent and structurally different inhibitors are therefore developed.
Ceritinib (LDK378, zykadia) is a potent ALK inhibitor compared to crizotinib [60][61][62]. A phase I study with 130 patients with ALK-positive advanced tumors included 122 NSCLC [63]. The doses were 50 to 750 mg in the dose escalation phase which enrolled 59 patients. The MTD of ceritinib was shown to be 750 mg daily. The dose-limiting toxicities (DLT) were diarrhea, vomiting, dehydration, elevated aminotransferase levels, and hypophosphatemia. Seventy-one patients received ceritinib in the dose expansion phase. One hundred fourteen patients received ceritinib dose of at least 400 mg daily. The ORR was 58 % (95 % CI 48-67). Among the 80 patients who failed crizotinib, the response rate was 56 % (95 % CI, 45 to 67). Among patients with NSCLC who received ceritinib with doses 400 mg or higher, the median PFS was 7.0 months (95 % CI, 5.6 to 9.5).
Thus, this study proved that ceritinib induced high responses in patients who failed crizotinib. Ceritinib was approved for treatment of relapsed or refractory NSCLC after crizotinib [64] (Table 1).

Alectinib
Alectinib (CH5424802, alecensa) is a potent and highly selective inhibitor of ALK tyrosine kinase with IC 50 of 1.9 nM [65,66]. More importantly, it has activity against L1196M which is one of the commonly seen secondary mutations in ALK gene leading to resistance to crizotinib.
In a multicenter, single-arm, open-label phase 1-2 study conducted in Japan (AF-001JP), ALK inhibitor naïve patients who had ALK-positive NSCLC were treated with alectinib [67]. In the dose escalation phase which included 24 such patients, increasing doses in the 20-300 mg range were used. No dose-limiting toxicities (DLT) were noted. Hence, 300 mg twice daily was established as the recommended dose for phase II. The phase II portion enrolled 46 patients. Forty-one of these patients had PR and 2 had CR. Hence, ORR was around 94 % (95 % CI: 82-98). Grade 3 adverse events were reported in 26 % (n = 12) and included elevated creatinine phosphokinase and neutropenia [67].
In another phase 1-2 single-arm open-label study, 47 patients with ALK-positive NSCLC who had resistance to crizotinib or were intolerant were treated with alectinib [68]. In the dose escalation phase, doses were escalated from 300 to 900 mg in seven different cohorts of patients. DLTs were seen in the 900-mg cohort: grade 3 headache in one patient and grade 3 neutropenia in another one. Three patients dropped out of the study due to adverse events: grade 3 dyspnea, grade 4 CNS metastasis, and grade 3 abdominal pain. Out of the 47 patients, 44 were assessed for response and the ORR was found to be 55 % (24 PR, one CR). ORR for the 21 patients who had baseline CNS metastasis was 52 % (5 CR and another 6 having partial CNS response). Therefore, this study showed that alectinib not only was effective in patients pretreated with first-generation ALK inhibitor but also was active for CNS metastasis [68]. Alectinib is now FDA-approved for the treatment of metastatic ALK + NSCLC in patients who have progressed on or are intolerant to crizotinib.
In the last update of the phase I/II single-arm, openlabel, multicenter study in patient pts with advanced malignancies (NCT01449461), patients received brigatinib as the following: phase I: 30-300 mg/day total daily dose; phase II: 90 mg/day, 180 mg/day, or 90 mg/day for 7 days followed by 180 mg/day. Safety was reported in all 137 treated pts; efficacy was evaluated in all 79 ALK+ NSCLC pts [72] (Table 2). Most common treatment-emergent adverse events (TEAE) included nausea, diarrhea, fatigue, cough, and headache. Early-onset
To overcome ALK mutations and ALK inhibitor resistance, lorlatinib was combined with PI3K pathway inhibitors, such as PF-05212384 (PI3K/mTOR), GDC0941 (pan-PI3K), or GDC0032 (beta-sparing). Such rational combination was reported to lead to more robust activity in vitro and greater duration of efficacy in vivo in the ALK inhibitor resistant models [76]. Lorlatinib is being studied in a phase I clinical trial in patients who were refractory to crizotinib and ceritinib (NCT01970865) [77]. One patient enrolled to this trial responded to lorlatinib for 8 months. Interestingly, the patient was resensitized to crizotinib after the patient failed the lorlatinib treatment, indicating that retreatment under molecular guidance can be a clinically meaningful approach.

Conclusions
Second-and third-generation ALK inhibitors are entering clinical applications for ALK+ NSCLC. Among these, dual inhibitors targeting ALK as well as EGFRm and ROS1 may provide additional benefits for crizotinibrefractory patients. Resensitization to and retreatment with crizotinib can be considered under molecular guidance. More and more biomarker-targeted agents are entering clinical applications [78][79][80][81]. Immune therapies are showing remarkable benefits [82][83][84][85][86][87][88][89][90][91]. It is foreseeable that combination of these novel agents and small molecular inhibitors may expand the potential for treatment of refractory lung cancer patients.