The treatment of patients with advanced non–small cell lung cancer with anaplastic lymphoma kinase chromosomal rearrangements has been revolutionized by the development of tyrosine kinase inhibitors (TKIs). Excellent progress has been made over the past decade, with 4 TKIs now approved in the front-line setting. Alectinib is the preferred first-line option based on its efficacy and side-effect profile. The central nervous system (CNS) activity of alectinib and brigatinib has allowed for treatment of CNS metastases with TKI therapy. Once resistance inevitably develops, newer therapies such as lorlatinib can be considered.
Key points
- •
Non–small cell lung cancer with anaplastic lymphoma kinase (ALK) chromosomal rearrangement is sensitive to treatment with tyrosine kinase inhibitors (TKIs).
- •
Numerous TKIs have been developed in recent years, including alectinib, which is the current preferred first-line agent for treatment-naïve patients.
- •
The development of resistance has led to next-generation ALK inhibitors that better penetrate the central nervous system, which has improved the treatment of brain metastasis.
Introduction
First described in 2007, echinoderm microtubule-associated protein-like 4 (EML4)-anaplastic lymphoma kinase (ALK) gene rearrangements are chromosomal inversions that lead to constitutive oncogenic activation. Cells harboring ALK rearrangements are sensitive to tyrosine kinase inhibitors (TKIs), which have dramatically improved patient outcomes, with up to 50% of patients surviving 6.8 years after diagnosis.
The proportion of patients with ALK-positive disease varies depending on the population. ALK rearrangements are detected in approximately 3% to 7% of adenocarcinomas, translating to 60,000 new cases annually worldwide. This incidence is similar in Asian (4.2%) and Western (3.4%) populations. ALK-positive disease occurs almost exclusively in adenocarcinomas, is associated with younger age (median age of diagnosis of 52), male sex, and never-smoking or light-smoking history.
ALK-positive patients with stage I–III disease are managed similarly to those with wild-type disease. They may receive surgery, radiation, chemotherapy, or multimodal therapy as appropriate for the patient’s stage. It remains to be determined whether TKIs will play a role for these early-stage patients (see later discussion). For patients with previously untreated metastatic ALK-positive disease, multiple TKIs are available ( Table 1 ). Drug resistance inevitably emerges, and has led to the development of more potent next-generation ALK inhibitors ( Table 2 ).
Drug, Trial Name | Phase | Arms | N | ORR (%) | ORR P Value | Median PFS (Months) | PFS P Value | OS (Months) | OS P Value |
---|---|---|---|---|---|---|---|---|---|
Crizotinib, PROFILE 1014 | 3 | Crizotinib Chemotherapy | 172 171 | 74 45 | <.001 | 10.9 7.0 | <.001 | NR 47.5 | .0978 |
Ceritinib, ASCEND 4 | 3 | Ceritinib Chemotherapy | 189 187 | 72.5 26.7 | <.01 | 16.6 8.1 | <.00001 | NR 26.2 | .056 |
Alectinib, J-ALEX | 3 | Alectinib Crizotinib | 103 104 | 92 79 | NS | 34.1 a 10.2 a | <.0001 | NR a NR a | NA |
Alectinib, ALEX | 3 | Alectinib Crizotinib | 152 151 | 82.9 b 75.5 b | .0936 | 34.8 b 10.9 b | <.001 | NA | NA |
Alectinib, ALESIA | 3 | Alectinib Crizotinib | 125 62 | 91 77 | <.01 | NR 11.1 | <.0001 | NA | NA |
Brigatinib, ALTA-1L | 3 | Brigatinib Crizotinib | 137 138 | 71 60 | NS | NR c 9.8 | NA | NA | NA |
Lorlatinib, CROWN | 3 | Lorlatinib Crizotinib | Ongoing | ||||||
Ensartinib, eXALT3 | 3 | Ensartinib Crizotinib | Ongoing |
b Follow-up study after additional 10 months, Camidge et al.
c 12-month PFS rate of 67% with brigatinib versus 43% for crizotinib, P <.001.
Drug, Trial Name | Phase | Prior Therapies | Arms | N | ORR (%) | ORR P Value | Median PFS (Months) | PFS P Value | OS a (Months, HR, or Rate) | OS P Value |
---|---|---|---|---|---|---|---|---|---|---|
Crizotinib, PROFILE 1007 | 3 | Platinum-based chemotherapy | Crizotinib Chemotherapy | 173 174 | 65 20 | <.001 | 7.7 3.0 | <.001 | 20.3 mo 22.8 mo | .54 |
Ceritinib, ASCEND 5 | 3 | Crizotinib, platinum-based chemotherapy | Ceritinib Chemotherapy | 115 116 | 39 7 | <.01 | 5.4 1.6 | <.01 | 1.0 b | .50 |
Alectinib, ALUR | 3 | Chemotherapy Crizotinib | Alectinib Chemotherapy | 72 35 | 37.5 2.9 | <.01 | 7.1 1.6 | <.001 | NR 12.6 mo | NS |
Brigatinib, ALTA , | 2 | Crizotinib | Low dose c High dose c | 112 110 | 48 d 53 d | NS | 9.2 d 16.7 d | NS | 29.5 mo e 34.1 mo e | NA |
Lorlatinib, Solomon et al, 2018 f | 2 | Crizotinib g >1 ALK g inhibitor >2 ALK g inhibitors | Lorlatinib | 59 198 111 | 70 47 39 | NA | NR 7.3 6.9 | NA | NA | NA |
ASCEND 9 | 2 | Alectinib ± crizotinib | Ceritinib | 20 | 25 | NA | 3.7 | NA | 75.6% h | NA |
a OS is reported as either absolute value in months, HR comparing groups, or rate after a specified amount of time.
c Low dose represented 90 mg daily, high dose represented 90 mg daily for 7 days followed by 180 mg daily.
d Longer-term data released after 8 months of follow-up.
e Longer-term data release after greater than 20 months of follow-up.
g Includes those with and without chemotherapy.
Diagnostic testing
All patients with metastatic lung adenocarcinoma should be tested for ALK rearrangements using fluorescence in situ hybridization (FISH), immunohistochemistry (IHC), or next-generation sequencing (NGS). FISH is the gold standard for detecting ALK rearrangements, and uses red and green probes to hybridize with either side of the ALK translocation breakpoint. These probes either overlay to form a yellow signal for wild-type samples or separate with independent red and green signals if a fusion mutation is present.
An alternative to FISH is IHC, which involves the use of monoclonal antibodies to the ALK fusion oncogene. NGS is performed with extraction of genomic DNA from tumor cells, with probes targeting cancer-specific genes using either plasma or tissue. Comparing these 3 methods, IHC had the greatest positive rate (94.5%), with similar results from NGS (92.7%), followed by FISH (82.4%).
Tyrosine kinase inhibitors
Crizotinib
Crizotinib was the first ALK TKI developed. PROFILE 1007 (N = 347) was a phase 3 trial that compared crizotinib with chemotherapy in patients previously treated with chemotherapy. Crizotinib demonstrated an improvement in objective response rate (ORR) (65% vs 20%, P <.001), progression-free survival (PFS) (7.7 months vs 3.0, P <.001), and quality of life. In the phase 3 study PROFILE 1014 (N = 343), which compared crizotinib with chemotherapy in the front-line setting, crizotinib demonstrated an improved ORR (74% vs 45%, P <.001) and PFS (10.9 vs 7.0 months, P <.001). In both trials, there was no overall survival (OS) difference, likely because of crossover. , These trials established the role of crizotinib as the first-line standard of care.
Toxicity
Gastrointestinal toxicity is the most common side effect of crizotinib, including diarrhea (any grade, 61%; grade 3 or 4%, 22%), vomiting (46%), and constipation (43%). Grade 3 or 4 elevated liver function tests (LFTs) was seen in 14% of patients ; LFTs should be monitored every 2 weeks for the first 2 months and periodically thereafter. Cardiac toxicity (including bradycardia and QTc prolongation) is also observed. Visual disturbances were seen in 71% of patients, but only 1% were grade 3 or higher.
Central nervous system metastasis
Crizotinib is largely ineffective at controlling central nervous system (CNS) disease, with an intracranial (IC) response rate as low as 7%. The CNS is the most common site of progression for patients on crizotinib. Crizotinib is a known substrate of the drug efflux pump p -glycoprotein, limiting its CNS accumulation. , This lack of CNS efficacy helped drive the development of several next-generation ALK inhibitors.
Ceritinib
Ceritinib is a second-generation ALK inhibitor with activity against IGF-R1, IR, and ROS-1. It has activity in crizotinib-resistant disease, targeting resistance mutations such as the common variants L119M and G1269A, as well as I1171T and S1206Y. Ceritinib also better penetrates the blood-brain barrier.
Ceritinib is approved in both treatment-naïve and crizotinib-resistant patients. ASCEND-5 (N = 231) demonstrated a clinical benefit for ceritinib over third-line chemotherapy in patients previously treated with both platinum-based chemotherapy and crizotinib. ASCEND-4 (N = 376) compared ceritinib with chemotherapy in treatment-naïve patients, including those with asymptomatic or stable brain metastases. Median PFS for the ceritinib group was 16.6 months versus 8.1 months in the chemotherapy group ( P <.00001). Neither ASCEND-4 nor ASCEND-5 demonstrated an OS benefit, likely because of high crossover. Poor tolerability of ceritinib and the clinical efficacy of alectinib (see later discussion) hindered ceritinib’s widespread use.
Whether there will be a role for ceritinib in the treatment of alectinib-resistant disease remains to be seen. ASCEND-9 (N = 20) studied ceritinib in patients who progressed on alectinib and demonstrated an ORR of 25% and disease control rate (DCR) of 70%.
Central nervous system metastasis
Ceritinib has excellent CNS activity with an IC response rate as high as 72%. , Despite this robust activity, the mean PFS for ceritinib-treated patients with brain metastases was still shorter than those without (10.7 vs 26.3 months). An ongoing phase 2 study ASCEND-7 is evaluating the efficacy and safety of ceritinib in patients with brain metastasis.
Toxicity
Grade 3 or 4 toxicities are seen in as many as 78% of patients. Gastrointestinal toxicity of any grade includes diarrhea (85%), nausea (69%), vomiting (66%), and abdominal pain (25%). More than 70% had grade 3 or worse LFT abnormalities, with notable anorexia and fatigue. Poor tolerability of the approved dose of ceritinib (750 mg once daily while fasting) hindered its adoption, although the new lower dose approved by the Food and Drug Administration (450 mg dose with food) has improved the side effect profile tolerability. ,
Alectinib
Alectinib is a second-generation ALK inhibitor that, owing to robust clinical efficacy and a superior safety profile, is the preferred first-line option for metastatic ALK-positive NSCLC. Alectinib is not a substrate of the p -glycoprotein efflux transporter, allowing it to effectively penetrate the CNS.
In the global phase 3 ALEX trial (N = 303), treatment-naïve patients were randomized to first-line alectinib or crizotinib. Median PFS with alectinib was 35 months versus 11 months with crizotinib (hazard ratio [HR] 0.43, P <.001), , although OS data await maturation. Similar results were seen in Japanese patients in J-ALEX (N = 207), as well as a third more recent phase 3 study ALESIA, which compared alectinib with crizotinib in Asian patients. These trials have led to alectinib as the preferred front-line treatment.
In addition, alectinib plays a role in the treatment of patients who progressed on or were intolerant to crizotinib. The phase 3 trial ALUR (N = 107) randomized patients previously treated with chemotherapy and crizotinib to alectinib or chemotherapy. Treatment with alectinib resulted in a PFS of 7.1 months versus 1.6 months in those treated with chemotherapy (HR 0.32, P <.01).
Central nervous system metastasis
Excellent CNS activity of alectinib has been consistently demonstrated across all trials. With ALEX, time to CNS progression was significantly longer in the alectinib group compared with the crizotinib group (HR 0.16, P <.001). For patients with metastatic CNS disease at baseline, IC responses were achieved in 75% (n = 16) of patients treated with alectinib. Therefore, many patients with CNS disease can be treated with TKIs alone without local therapy (surgery or radiation).
Toxicity
Alectinib is tolerated much better than crizotinib, with grade 3 to 5 toxicities in approximately 40% of patients. , , The side-effect profile includes nausea (14%), diarrhea (12%), vomiting (7%), and elevated bilirubin (15%). Alectinib can also cause myalgias (16%), and for this reason creatine kinase (CK) levels are tested every 2 weeks in the first month of therapy. Anemia (20%) and photosensitivity (5%) are also reported.
Brigatinib
Brigatinib is a second-generation oral TKI that can overcome several crizotinib resistance mutations. Brigatinib was granted accelerated approval for treatment of ALK-positive NSCLC in patients who have progressed on crizotinib, with ongoing trials in the front-line setting.
The ALTA-1 trial randomized 275 treatment-naïve patients to brigatinib or crizotinib. PFS at 12 months was improved with brigatinib (67% vs 43%, HR 0.49, P <.001). OS in the front-line setting has not yet been released. There are no trials comparing brigatinib with alectinib in the treatment-naïve setting. Brigatinib has showed some promise in crizotinib-refractory patients with PFS as high as 16.7 months in a recent phase 2 trial. There are limited data supporting the use of brigatinib at time of progression on alectinib, with PFS from 4.4 to 6.6 months.
Central nervous system metastasis
Brigatinib has excellent IC activity, with 78% (n = 18) of patients with measurable brain metastases at baseline demonstrating an objective IC response (ICR) in the front-line setting. Updated phase 2 results demonstrated that crizotinib-refractory patients with measurable CNS lesions at baseline had a nearly 67% (n = 18 patients) ICR with a median IC PFS of 18.4 months.
Toxicity
In ALTA-1L, grade 3 or higher adverse events occurred in 61% of patients treated with brigatinib (n = 136). Similar to the other ALK inhibitors, the most common adverse events of any grade were gastrointestinal (49%), elevated CK (39%), and increased alanine aminotransferase ALT (19%). Amylase and lipase elevation was also observed (any grade rates of 19% and 14%, respectively), as was cardiac toxicity (particularly bradycardia). Blood pressure should be controlled before brigatinib initiation and monitored monthly, because hypertension has also been observed (n = 136, 23%).
A distinguishing feature of brigatinib is the potential for severe pulmonary toxicity. This typically occurs within 24 to 48 hours of initiation, and manifests as dyspnea and hypoxia with ground-glass opacification and interstitial opacities on imaging. This occurs in 3% to 6% of patients, , , appears to be more common in patients with prior crizotinib treatment, and is dose related. Step-up dosing over 7 days is used to mitigate the risk of early pulmonary events. , For patients with new or worsening respiratory symptoms, brigatinib should be held with prompt evaluation for interstitial lung disease or pneumonitis, and if found to be grade 3 or higher, brigatinib should be discontinued. ,
Lorlatinib
Lorlatinib is a third-generation selective inhibitor of ALK and ROS1. Shaw and colleagues have demonstrated an ORR of 57% (n = 28) for patients harboring the G1202R mutation, commonly found after second-generation ALK inhibitors.
A recent phase 2 study by Solomon and colleagues enrolled 228 ALK-positive patients with a range of prior treatment exposures. Clinical performance varied based on prior TKI exposure. Crizotinib-only patients demonstrated an ORR of 69% (n = 228). Median PFS was not reached. Patients who had failed 2 or more ALK TKIs had a less robust ORR of 39% with PFS of 6.9 months.
An ongoing phase 3 study (“CROWN”) is randomizing treatment-naïve ALK-positive NSCLC to either lorlatinib or crizotinib in the first-line setting. A French study, LORLATU (NCT: 02327477) is evaluating treatment sequences in patients receiving lorlatinib. There have not been any studies comparing lorlatinib with chemotherapy in alectinib-resistant disease.
Central nervous system metastasis
The mean cerebrospinal fluid to plasma concentration ratio for lorlatinib was 0.75 in early trials, confirming significant CNS penetration. In patients with baseline brain metastasis previously on at least one ALK inhibitor, lorlatinib had an IC response rate of 63% (n = 81) with median duration of IC response at 14.5 months.
Toxicity
Grade 3 to 4 hyperlipidemia is significant (31%), with 81% of patients requiring a lipid-lowering agent. Peripheral edema of any grade was present in 43% of patients, with peripheral neuropathy in 30%. CNS effects of any grade were reported in 39% of patients including change in cognition (23%), mood (22%), and speech (8%). Most cognitive effects were found to be mild and were rapidly reversible with dose reduction.
Tyrosine kinase inhibitors in development
Ensartinib (X-396) is a novel inhibitor with activity against ALK and ROS1 crizotinib resistance mutations, including L1196M and C1156Y. In a recent phase 1/2 study (N = 60), relative response (RR) rate was 60% with a median PFS of 26.2 months. Activity was seen in patients who had previously received a second-generation ALK TKI (n = 16, ORR 23%, DCR 50%), as well as in patients who had received between 2 and 5 prior regimens of ALK TKIs. Intracranial RR (64%) and IC DCR (92.9%) were particularly promising. A phase 3 trial, eXalt3, is currently comparing ensartinib with crizotinib.
Entrectinib is another novel inhibitor of ALK. A recent phase 1 study found that entrectinib had rapid and durable benefit in patients harboring rearrangements of NTRK, ROS1, and ALK. However, no responses were observed in patients previously treated with ALK inhibitors.
Repotrectinib is a next-generation ROS, TRK-A, and ALK inhibitor that has shown meaningful preliminary clinical activity. The current phase 1/2 trial (Trident-1) is still ongoing.
Resistance mechanisms
Mutations in the ALK kinase domain are the best described mechanism of TKI resistance, occurring in both crizotinib-treated patients (20%–36%) , and patients treated with a second-generation ALK inhibitor (>50%). Each TKI is associated with an individual spectrum of mutations. , Gainor and colleagues analyzed 103 biopsies from ALK-positive patients progressing on various ALK inhibitors. L1196M is the most common mutation found in crizotinib resistance (7%). G1202R occurs frequently after second-generation agents, including ceritinib (21%), alectinib (29%), and brigatinib (43%). , , , Even though G1202R confers high-level resistance, it can be overcome by lorlatinib. , Little is known regarding lorlatinib resistance.
Resistance mechanisms outside of the ALK kinase domain also occur. Amplification of the ALK domain occurs in about 10% of crizotinib-resistant samples, either in isolation or with other mutations. , Bypass signaling pathway activation is another resistance mechanism, with activation of the epidermal growth factor receptor (EGFR) most commonly reported. , Transformation to both spindle-cell morphology and squamous cell lung cancer have also been described. ,
Systemic therapy after tyrosine kinase inhibitors
Treatment with chemotherapy, with or without immunotherapy, is used after targeted therapies have been exhausted. The use of combination chemoimmunotherapy is supported by a subgroup analysis of 111 patients with EGFR and ALK mutations from the IMpower150 study. The combination of carboplatin, paclitaxel, bevacizumab, and atezolizumab demonstrated an improvement of PFS in patients with EGFR and ALK mutations over carboplatin, paclitaxel, and bevacizumab (9.7 months vs 6.1 months; HR 0.59, 95% confidence interval 0.37–0.94).
Outside of IMpower150, data supporting the use of immunotherapy in ALK-mutated patients is scarce. Patients with ALK mutations were excluded from KEYNOTE-189, which studied the combination of carboplatin, pemetrexed, and pembrolizumab. Negative clinical trial results were seen for ALK patients with carboplatin/nab-paclitaxel/atezolizumab, , nivolumab, pembrolizumab, and atezolizumab. In addition, extrapolation of EGFR data lead to safety concerns in giving TKI therapy either with or following immunotherapy. Testing for ALK should be completed before the administration of chemoimmunotherapy in the untreated setting to avoid this increased toxicity risk.
Discussion
Alectinib is the current preferred first-line treatment for ALK-positive, metastatic NSCLC. This preference is based on improved efficacy, side-effect profile, and CNS activity when compared with crizotinib. , Brigatinib, ceritinib, and crizotinib remain options for front-line treatment, and may be considered based on individual clinical circumstances. Lorlatinib should be considered after progression on alectinib.
Excellent CNS activity of alectinib and brigatinib has allowed for treatment of CNS metastasis with TKI therapy and has allowed patients to delay or avoid the use of radiotherapy or surgery for local control. , , , , For patients with CNS progression on alectinib, lorlatinib remains an option.
Future directions
Long-term follow-up data for brigatinib await maturity, and how this drug will compare with alectinib remains to be seen. Ensartinib, entrectinib, and repotrectinib are all being investigated in clinical trials, and the role they will play is yet to be determined. Additional data are needed regarding the optimal treatment following TKI therapy, whether it be chemotherapy or chemoimmunotherapy.
The role of TKIs in the perioperative setting is under investigation. Crizotinib is being studied in the neoadjuvant setting as well as the adjuvant setting as part of the larger, multi-institutional Adjuvant Lung Cancer Enrichment Marker Identification and Sequencing Trial (ALCHEMIST). , An ongoing phase 3 study (ALINA) is evaluating the efficacy and safety of alectinib compared with platinum-based chemotherapy in patients with completed resected IB to stage IIIA tumors as an adjuvant therapy.
In summary, excellent progress has been made over the past decade in the treatment of ALK-positive NSCLC, with 4 TKIs now approved in the front-line setting, with additional options available at time of progression. Next-generation ALK inhibitors demonstrate excellent CNS response rates. Future areas of research include additional next-generation TKIs, the perioperative role of TKIs, and the optimal systemic treatment once TKI options have been exhausted.
Disclosure
Dr A.V. Serritella has nothing to disclose. Dr C.M. Bestvina discloses the following: Consulting: AbbVie, AstraZeneca, Genentech, Pfizer. Honorarium: OncLive. Travel: EMD Serono.