Anaplastic lymphoma kinase (ALK) fusion oncogene positive non-small cell lung cancer

INTRODUCTION — Anaplastic lymphoma kinase (ALK) is a tyrosine kinase that can be aberrantly expressed in several tumor types. In non-small cell lung cancer (NSCLC), chromosomal rearrangements involving the ALK gene loci on chromosome 2 are found in approximately 3 to 5 percent of NSCLC tumors [1,2]. The most common ALK rearrangement in NSCLC juxtaposes the 5' end of the echinoderm microtubule-associated protein-like 4 (EML4) gene with the 3' end of the ALK gene, resulting in the novel fusion oncogene EML4-ALK [3]. This fusion oncogene is transforming both in vitro and in vivo and defines a distinct clinicopathologic subset of NSCLC. Detecting ALK gene rearrangements in newly diagnosed patients with advanced/metastatic/recurrent NSCLC is critical, as the presence of this oncogene influences treatment decisions. (See 'Treatment approach' below.)

Tumors that contain ALK fusion oncogenes or its variants are associated with specific clinical features, including never- or light-smoking history and younger age, although these "typical characteristics" are not absolutely required to be fulfilled for a diagnosis of ALK-rearranged lung cancer. Testing for this fusion gene in NSCLC is important, as "ALK-positive" tumors (tumors harboring a rearranged ALK gene/fusion protein) are highly sensitive to therapy with ALK-targeted inhibitors.

The molecular pathogenesis, clinical features, and treatment of NSCLC associated with the ALK fusion oncogene are discussed here.

An overview of the treatment of metastatic NSCLC is presented separately, as are the methods and indications for molecular testing. (See "Overview of the initial treatment of advanced non-small cell lung cancer" and "Personalized, genotype-directed therapy for advanced non-small cell lung cancer", section on 'Molecular testing'.)

DIAGNOSIS — We obtain molecular testing in all patients with advanced or metastatic nonsquamous NSCLC, regardless of smoking history, including assessment for ALK gene rearrangements. This is expected standard of clinical care. ALK gene rearrangements may be detected in tumor or plasma specimens [4]. Available methods include next-generation sequencing (NGS), fluorescence in situ hybridization (FISH), immunohistochemistry (IHC), and reverse transcription polymerase chain reaction [5,6].

In the United States, given the number of US Food and Drug Administration (FDA)-approved therapeutically actionable biomarkers in NSCLC, we perform NGS assays, which can simultaneously assess multiple biomarkers concurrently. Other methods of affirming ALK positivity include FISH using the FDA-approved test (Vysis Probes) and IHC using the FDA-approved Ventana ALK (D5F3) companion diagnostic (CDx) assay. In Europe, IHC is widely used to detect ALK protein expression within a tumor sample.

●NGS – Multigene NGS assays have rapidly emerged as a preferred option over either FISH or IHC. In a single-center study of 55 patients with ALK-positive NSCLC as assessed by either FISH, IHC, or NGS (most of whom had all three tests concurrently), the positive rate with IHC was 95 percent, followed by NGS at 93 percent, and FISH at 82 percent, among which IHC and NGS had the highest concordance rate of 87 percent [7]. In a separate study, comprehensive genomic profiling detected ALK rearrangements in a subset of patients with NSCLC with negative ALK FISH results [8]. These patients had responses to ALK inhibitors, comparable to historic response rates of patients with ALK FISH-positive tumors.

●FISH – FISH was the first clinical test widely utilized to detect ALK rearrangements in NSCLC [9-12]. The commercial break-apart probes include two differently colored (red and green) probes that flank the highly conserved translocation breakpoints within ALK (picture 1 and picture 2). In nonrearranged cells, the overlying red and green probes result in a yellow (fused) signal; in the setting of an ALK rearrangement, these probes are separated, and splitting of the red and green signals is observed. Atypical patterns of rearrangement have also been identified, and these are also responsive to ALK inhibition [13]. ALK gene amplification alone is not predictive of responsiveness to these agents and does not carry the same significance as rearrangement.

●IHC – Multiple monoclonal antibodies have been developed for the IHC detection of the ALK fusion oncoprotein, and IHC using these antibodies is highly sensitive and specific [14]. Thus, IHC is an appropriate method to screen for and identify the presence of ALK positivity, and the Ventana ALK (D5F3) CDx assay has been approved for use in the United States by the FDA [15].

CLINICOPATHOLOGIC FEATURES — With increasing identification of this molecular abnormality, the key epidemiologic, demographic, and pathologic features associated with ALK fusion oncogenes have been identified.

●Epidemiology – In unselected NSCLC populations, the ALK rearrangement is detected in approximately 3 to 5 percent [1,10-12,16-21].

Brain metastases are more common among patients with ALK-positive NSCLC compared with patients with NSCLC lacking an epidermal growth factor receptor (EGFR) or ALK driver. (See "Brain metastases in non-small cell lung cancer", section on 'Epidemiology of brain metastases in NSCLC'.)

●Age of onset – Patients with ALK-positive lung cancer are relatively younger at the time of diagnosis compared with unselected populations of patients with advanced NSCLC [11,22]. The two studies that were used to support the approval of crizotinib included 255 patients whose tumors were ALK positive; in this database, the median age was 52 years (range, 21 to 82 years) [22]. The estimated median age for unselected patients with lung cancer is approximately 70 years [23].

Interestingly, other cancers known to harbor ALK rearrangements, such as nucleophosmin (NPM)-ALK-positive anaplastic large cell lymphoma, are also associated with younger age and are, in fact, most common in children and young adults.

●Smoking history – The ALK fusion oncogene in patients with NSCLC is typically, but not exclusively, associated with a history of never or light smoking (<10 pack-years) [3,11,22]. In the crizotinib study database of 255 patients, never-smokers and former smokers comprised 70 and 28 percent of cases, respectively [3,11,22].

●Histology – The majority of lung tumors that harbor the ALK fusion oncogene are adenocarcinomas. In the 255 patients with the ALK fusion oncogene included in the crizotinib database, 97 percent had adenocarcinoma [22]. ALK rearrangement has been reported in squamous cell carcinoma, but this is rare [10,20]. In terms of the specific type of adenocarcinoma, some studies have reported that ALK-positive tumors are more likely to have abundant signet ring cells than those with an EGFR mutation or wild-type tumors [24].

TREATMENT APPROACH

Rationale for ALK inhibitors — Lung tumors that harbor ALK rearrangements are sensitive to ALK tyrosine kinase inhibitors (TKIs). Treatment with ALK TKIs should be limited to patients whose tumors contain this abnormality as demonstrated by fluorescence in situ hybridization, next-generation sequencing, or immunohistochemistry. Patients should have tumor tissue assessed not only for the presence of ALK rearrangement, but also for other oncogenic driver mutations that are therapeutically targetable in NSCLC. (See "Personalized, genotype-directed therapy for advanced non-small cell lung cancer", section on 'Molecular testing'.)

An ALK inhibitor is preferred as the initial therapy for patients whose tumor contains this genetic abnormality. Results of a phase III trial comparing ALK inhibition using the first-generation ALK TKI, crizotinib, with chemotherapy in treatment-naïve patients demonstrated a prolongation in progression-free survival (PFS) and improved response rate and quality of life [25]. No significant differences in overall survival (OS) were seen (hazard ratio [HR] 0.76, 95% CI 0.55-1.05), potentially due to the confounding effects of crossover, although an unprecedented median OS in excess of four years was reported in the crizotinib arm.

Subsequent trials have shown PFS benefits of second-generation ALK inhibitors including alectinib, brigatinib, and ensartinib over crizotinib [26-31]. A meta-analysis of six trials showed a statistically significant OS benefit of ALK inhibitors over chemotherapy (HR 0.84, 95% CI 0.72-0.97), despite most of the included trials having substantial crossover from chemotherapy to ALK inhibitors during the study period [32]. Moreover, ALK inhibitors increased the health-related quality of life measure and time to clinical deterioration. (See 'Preferred options' below and 'Crizotinib' below.)

●Approach for cancers that require urgent treatment, prior to results of genotype testing – If urgent systemic treatment is required before the results of genotype testing are available, systemic chemotherapy may be initiated. (See "Initial management of advanced non-small cell lung cancer lacking a driver mutation", section on 'PD-L1 low (<50 percent) or unselected tumors'.)

When the results of genotype testing become available, the treatment plan should be reassessed. There are no clinical trials that directly address the optimal timing of ALK inhibitors in patients who have already started on chemotherapy. For such patients, we prefer to switch to an ALK inhibitor when genotyping results identify ALK-positive tumors.

Efficacy of other approaches

●Chemotherapy – When patients with ALK-positive advanced NSCLC require chemotherapy, pemetrexed or a pemetrexed-based regimen is generally preferred, since almost all of these patients have nonsquamous histology. Pemetrexed-based chemotherapy appears to have slightly increased activity in ALK-positive NSCLC patients compared with other chemotherapy regimens.

●Immunotherapy – Early- and later-phase trials with single-agent programmed cell death protein 1 (PD-1) inhibitors suggest lower response rates among never-smoking patients, including those with ALK and epidermal growth factor receptor (EGFR) genetic aberrations, and subsequent trials have frequently excluded this subset of patients. Nevertheless, we consider immunotherapy in combination with chemotherapy (for example, as part of the IMPower150 regimen [33]), or as monotherapy in patients who have progressed on chemotherapy as well as available ALK inhibitors.

We do not combine immunotherapy with ALK inhibitors except on clinical trials, with close monitoring. Fatal hepatotoxicity has been reported with the combination of nivolumab and crizotinib [34]. (See "Initial management of advanced non-small cell lung cancer lacking a driver mutation".)

FIRST-LINE TREATMENT

Preferred options — For those with newly diagnosed, ALK-positive NSCLC, we recommend a next-generation ALK inhibitor as first-line treatment rather than crizotinib. In choosing between second-generation inhibitors, we suggest alectinib, given the benefit of longer-term follow-up of clinical trials with this agent compared with others. (See 'Alectinib' below.)

However, direct comparisons between second-generation inhibitors have not been performed, and as such, we recognize brigatinib or lorlatinib are other acceptable front-line options. (See 'Brigatinib' below.)

Alectinib — Alectinib is approved by the US Food and Drug Administration (FDA) for the first-line treatment of patients with ALK-positive metastatic NSCLC as detected by an FDA-approved test [35]. It is also approved for those who have progressed on crizotinib. Discussion of results in the first-line setting is found here, while data related to its use in the second-line setting and among patients with brain metastases are discussed elsewhere. (See "Brain metastases in non-small cell lung cancer", section on 'ALK translocations'.)

In a global study of 303 patients randomly assigned to first-line alectinib versus crizotinib (ALEX), those receiving alectinib had a reduction in risk of progression or death of 53 percent (hazard ratio [HR] 0.47, 95% CI 0.34-0.65), with median progression-free survival (PFS) not reached versus 11.1 months for those receiving crizotinib at a median follow-up of approximately 18 months [27]. In an update to the global ALEX study, with additional follow-up, median PFS was 35 months in the alectinib group versus 11 months in the crizotinib group (HR 0.43) [36]. Median overall survival (OS) was not reached with alectinib versus 57 months with crizotinib (stratified HR 0.67, 95% CI 0.46-0.98), but results were not yet mature.

The time to central nervous system (CNS) progression in the overall population was improved with alectinib (HR 0.16, 95% CI 0.10-0.28) [27]. Grade 3 to 5 toxicities were less frequent with alectinib (41 versus 50 percent). Adverse events that occurred at a higher incidence with alectinib than with crizotinib included anemia (20 versus 5 percent), myalgia (16 versus 2 percent), increased blood bilirubin (15 versus 1 percent), increased weight (10 versus 0 percent), and photosensitivity reaction (5 versus 0 percent). Adverse events that were more common with crizotinib included nausea (48 versus 14 percent with alectinib), diarrhea (45 versus 12 percent), and vomiting (38 versus 7 percent). (See 'Management of toxicities associated with ALK inhibitors' below.)

These results are similar to an earlier randomized trial conducted in Japan (J-ALEX), in which the PFS with not reached with alectinib and was 10.2 months with crizotinib (HR 0.34, 99.7% CI 0.17-0.70) [37]. In a third phase III study (ALESIA), alectinib was compared with crizotinib in Eastern Asian patients with untreated, advanced, ALK-positive NSCLC. A total of 187 patients were randomly assigned in a 2:1 ratio to receive either alectinib or crizotinib [38]. Median duration of follow-up was 15 to 16 months. Similar to J-ALEX and ALEX, alectinib was associated with a reduction in the risk of progression or death (HR 0.22, 95% CI 0.13-0.38), with median PFS not reached in the alectinib group versus 11.1 months in the crizotinib group.

Brigatinib — Brigatinib is a next-generation ALK inhibitor that targets a broad range of ALK mutations [39-42], and has demonstrated improved efficacy over crizotinib in the first-line treatment of ALK-positive, advanced NSCLC [29]. Brigatinib received FDA approval as a first-line treatment option for patients with advanced (stage IV) NSCLC based on the phase III ALTA 1L study [42]. It has shown comparable activity to alectinib in the first-line setting, but direct comparisons are lacking. Although brigatinib is associated with pulmonary toxicity in a small percentage of patients, this risk can be ameliorated by step-up dosing (90 mg once daily for seven days; if tolerated, then increase dose to 180 mg once daily).

In a phase III trial including 275 patients with advanced, treatment-naïve, ALK-positive NSCLC, those randomly assigned to brigatinib experienced an improvement in PFS over those assigned to crizotinib (12-month PFS rate of 67 versus 43 percent for brigatinib versus crizotinib, respectively [HR 0.49, 95% CI 0.33-0.74], at a median follow-up of between 9 and 11 months [29]; median PFS of 24 versus 11 months, respectively [HR 0.48, 95% CI 0.35-0.66 [43]). Even greater benefits were observed among those with baseline brain metastases. (See "Brain metastases in non-small cell lung cancer", section on 'Brain metastases at presentation'.)

Objective response rates (ORRs) were 79 percent with brigatinib versus 75 percent with crizotinib. Among those with brain metastases at baseline, the intracranial response rate was significantly higher with brigatinib (78 versus 26 percent). Further details of the intracranial efficacy of this agent in the first-line setting are discussed elsewhere. (See "Brain metastases in non-small cell lung cancer", section on 'Brain metastases at presentation'.)

Adverse events that occurred at a higher incidence with brigatinib than crizotinib included an increased creatine kinase level (39 versus 15 percent), cough (25 versus 16 percent), and hypertension (23 versus 7 percent). Symptoms possibly related to increased creatine kinase levels (myalgia and muscle pain) did not differ significantly between the groups. Interstitial lung disease (ILD)/pneumonitis at any time occurred in 4 percent of patients in the brigatinib group and 2 percent of patients in the crizotinib group. Grade 3 or 4 ILD/pneumonitis occurred in 3 and 0.7 percent, respectively. Adverse events that were more common with crizotinib than with brigatinib included nausea (56 versus 26 percent), diarrhea (55 versus 49 percent), constipation (42 versus 15 percent), peripheral edema (39 versus 4 percent), an increased alanine transaminase level (32 versus 19 percent), and visual impairment (16 versus 0 percent). Grade ≥3 adverse events occurred in 61 percent of patients receiving brigatinib and in 55 percent of patients receiving crizotinib.

Lorlatinib — Lorlatinib has regulatory approval in the front-line setting [44], and may be considered another option for first-line treatment.

In the phase III CROWN trial, treatment-naïve patients with ALK-positive stage IIIB/IV NSCLC were randomly assigned to lorlatinib or crizotinib. Among 296 patients, lorlatinib improved PFS relative to crizotinib [45,46]. At first interim analysis on the trial (approximately 18 months of follow-up), the median PFS for lorlatinib was not evaluable versus 9.3 months for crizotinib (HR 0.28 by blinded independent central review, 95% CI 0.19-0.41) [45]. Lorlatinib also showed strong intracranial activity. Grade 3 to 4 adverse events occurred in 72 percent receiving lorlatinib and 56 percent receiving crizotinib. Hypercholesterolemia and hypertriglyceridemia were observed in >70 percent of patients as were neurocognitive effects, which may impact the use of lorlatinib as a first-line option. (See "Brain metastases in non-small cell lung cancer", section on 'Brain metastases at presentation'.)

Less preferred options

Ceritinib — Ceritinib is another option for patients with advanced ALK-positive NSCLC, and has shown superiority over chemotherapy, but is less preferred than alectinib or brigatinib due to lesser efficacy in cross-trial comparisons. Ceritinib is a second-generation tyrosine kinase inhibitor (TKI) of ALK that is approximately 20 times more potent than crizotinib. Ceritinib is approved by the FDA for patients with metastatic NSCLC whose tumors are ALK positive as detected by an FDA-approved test [47]. While initial trials have been conducted using a fasting dose of 750 mg daily, a randomized, open-label trial demonstrated equivalence between this dose and 450 mg daily with food, and the lower dose is associated with decreased gastrointestinal toxicity [48]. The approved dose of ceritinib is 450 mg orally once daily with food [47].

Ceritinib has demonstrated improved efficacy over combination chemotherapy in the first-line setting. In the ASCEND-4 trial, which included 376 treatment-naïve, ALK-positive NSCLC patients, those randomly assigned to ceritinib (750 mg/day) experienced improved PFS (16.6 versus 8.1 months; HR 0.55, 95% CI 0.42-0.73), ORR (72.5 versus 26.7 percent), and duration of response (23.9 versus 11.1 months) compared with those assigned to pemetrexed and a platinum agent [30]. Results regarding intracranial efficacy of ceritinib from this study are discussed elsewhere. (See "Brain metastases in non-small cell lung cancer", section on 'ALK translocations'.)

In studies, the most frequent toxicities associated with ceritinib have been diarrhea, nausea, and vomiting, affecting the majority of patients, although adverse effects lead to ceritinib discontinuation in only approximately 5 percent [49,50]. (See 'Management of toxicities associated with ALK inhibitors' below.)

Crizotinib — Although crizotinib is more effective than chemotherapy for ALK-positive NSCLC, it should only be used if a next-generation ALK inhibitor is not available.

Crizotinib is a multitargeted TKI. It was the first ALK inhibitor used clinically and has demonstrated markedly improved outcomes in patients with ALK-positive advanced NSCLC relative to chemotherapy, both in the first-line and subsequent-line settings [25,51,52]. However, newer, more potent ALK inhibitors with greater systemic and CNS efficacy have since been developed and are preferred over crizotinib. (See 'Preferred options' above and 'Ceritinib' above.)

As newer ALK inhibitors are being used in the first-line setting only more recently, many patients will have been treated initially with and have developed resistance to crizotinib, making second-line ALK inhibitor treatment an important consideration. (See 'Treatment after progression on crizotinib' below.)

The efficacy of crizotinib has been demonstrated in randomized trials limited to patients whose tumors had the ALK rearrangement:

●Chemotherapy-naïve patients − In a randomized trial, 343 chemotherapy-naïve patients were assigned to crizotinib or chemotherapy with pemetrexed plus either cisplatin or carboplatin [25,52]. Crossover to crizotinib was permitted for those treated with chemotherapy. At a median follow-up of 17 months, PFS, the primary endpoint of the trial, was prolonged with crizotinib compared with chemotherapy (median, 10.9 versus 7 months; HR 0.45, 95% CI 0.35-0.60). The ORR was also increased (74 versus 45 percent). In the final analysis of the trial at 46 months' follow-up, the difference in OS was not significant (HR 0.76, 95% CI 0.55-1.05) [52]. However, 84 percent of patients assigned to initial chemotherapy subsequently were treated with crizotinib. After crossover adjustment, crizotinib improved OS over chemotherapy (HR 0.35, 95% bootstrap CI 0.08-0.72). Outcomes related to intracranial disease control are discussed below. (See 'Brain metastases' below.)

●Patients previously treated with chemotherapy − A phase III trial randomly assigned 347 patients who had previously been treated with one prior platinum-based chemotherapy regimen to either crizotinib or single-agent pemetrexed or docetaxel [51]. Patients who were assigned to chemotherapy were allowed to be treated with crizotinib when they developed progressive disease.

•At a median follow-up of 12 months, PFS, the primary endpoint of the trial, was increased with crizotinib compared with chemotherapy (median, 7.7 versus 3 months; HR for progression 0.49, 95% CI 0.37-0.64).

•The ORR based upon independent radiologic review was also increased (65 versus 20 percent). Responses were achieved more rapidly than with chemotherapy (median time to response, 6.3 versus 12.6 weeks) and were of longer duration (32 versus 24 weeks).

•There was no difference in OS (median, 20.3 versus 22.8 months; HR for death 1.02). The absence of an OS benefit presumably reflects subsequent treatment since 64 percent of chemotherapy-treated patients had crossed over to crizotinib after progressing on chemotherapy. (See 'First-line treatment' above.)

Options under investigation in the first-line setting — The next-generation inhibitor ensartinib (second-generation ALK TKI) has been evaluated against crizotinib in treatment-naïve patients with ALK-positive NSCLC, with promising results as first-line systemic therapy. However, OS results are not available. Ensartinib is not FDA approved and only available through clinical trials.

●Ensartinib – In eXalt3, among 290 patients with ALK-positive NSCLC, those assigned to ensartinib experienced an improvement in median PFS relative to those assigned to crizotinib (26 versus 13 months, respectively; HR 0.51, 95% CI 0.35-0.72) [31]. The most common treatment-related adverse events with ensartinib were rash (all grade approximately 70 percent, grade 1/2 approximately 60 percent) and transaminitis.

Duration of treatment — Treatment with ALK inhibitors is generally continued until there is evidence of disease progression. In carefully selected patients (eg, an isolated site of recurrence that can be treated with local therapy, those with extremely mild and asymptomatic progression), an ALK inhibitor may be continued after initial evidence of progressive disease [53]. Upon progression, treatment with a more potent next-generation ALK inhibitor or with standard chemotherapy may be indicated.

TREATMENT AFTER PROGRESSION ON CRIZOTINIB

Approach and evidence — Although a second-generation ALK inhibitor is now standard first-line treatment for patients with ALK-positive NSCLC, previously, crizotinib was used as first-line therapy, and there may be some patients who have only been treated with this agent. For ALK-positive patients who develop resistance to crizotinib or who are unable to tolerate crizotinib, we recommend treatment with one of the second-generation ALK inhibitors: alectinib or brigatinib.

Our preference in this setting is for alectinib or brigatinib, given the central nervous system (CNS) and systemic efficacy as well as tolerability.

Supporting data are as follows:

●Alectinib – In the phase III ALUR trial, 107 patients with advanced, ALK-positive NSCLC previously treated with platinum-based doublet chemotherapy and crizotinib were randomly assigned to alectinib or single-agent chemotherapy (pemetrexed or docetaxel) [54]. Median progression-free survival (PFS) was improved with alectinib (7.1 versus 1.6 months with chemotherapy; hazard ratio [HR] 0.32, 95% CI 0.17-0.59), and grade ≥3 adverse events were less frequent (27 versus 41 percent). CNS efficacy was also improved with alectinib, which is discussed elsewhere. (See "Brain metastases in non-small cell lung cancer", section on 'ALK translocations'.)

The efficacy of alectinib in the setting of crizotinib-resistant disease is also supported by two phase II trials [55,56]. In an analysis of the combined database from these two studies, there were 97 partial responses in 189 cases (51 percent) as determined by an independent review committee. In addition, there were 52 patients with stable disease (28 percent) [57]. The median duration of response was 15 months.

●Brigatinib – A phase II study of 222 patients with crizotinib-refractory, ALK-positive NSCLC demonstrated PFS of 9.2 and 16.7 months among those receiving a lower and higher dose of the agent, respectively, with low incidence of grade ≥3 toxicities in both arms [58]. Median overall survival (OS) was 29.5 months (18.2 months to not reached) versus 34.1 months (27.7 months to not reached). Objective response rates (ORRs) were 51 and 56 percent. However, brigatinib has been associated with early pulmonary toxicity, which occurred in 6 percent in this phase II trial, approximately one-half of which were grade ≥3 [59]. (See 'Management of toxicities associated with ALK inhibitors' below.)

Intracranial efficacy is discussed in detail elsewhere. (See "Brain metastases in non-small cell lung cancer", section on 'Crizotinib-resistant setting'.)

A less preferred option is ceritinib, which shows improved outcomes relative to chemotherapy, but cross-trial comparisons suggest lesser efficacy than the agents discussed above. A randomized phase III study demonstrated that, among ALK-positive NSCLC patients who have progressed on crizotinib, ceritinib yields improved outcomes relative to single-agent chemotherapy. In the open-label ASCEND-5 study, in which 231 patients who had received crizotinib were randomly assigned to ceritinib 750 mg/day or chemotherapy, those receiving ceritinib experienced improved PFS (5.4 versus 1.6 months; HR 0.49) and ORR (39.1 versus 6.9 percent), differences that were both statistically significant [60]. Although no improvement in OS was noted among those assigned to ceritinib, the OS analysis is immature. Furthermore, 75 patients assigned to chemotherapy crossed over to ceritinib upon progression, potentially diluting an OS benefit. While initial trials have been conducted using a fasting dose of 750 mg daily, a randomized, open-label trial demonstrated equivalence between this dose and 450 mg daily with food, and the lower dose is associated with decreased gastrointestinal toxicity [48,61]. The US Food and Drug Administration-approved dose of ceritinib is 450 mg orally once daily with food [47], although in other parts of the world (eg, the European Union, United Kingdom, and Canada), the approved dose is 750 mg, taken in a fasting state [62-64].

Next-generation inhibitors have not been compared head to head in crizotinib-resistant, ALK-positive lung cancer, and we recognize the limitations of cross-trial comparisons. An ongoing study (ALTA-3) is comparing alectinib and brigatinib in this setting.

Resistance mechanisms — Several distinct mechanisms of resistance to crizotinib have been reported in the literature, including:

●In approximately one-third of resistant cases, tumors have acquired a secondary mutation within the ALK tyrosine kinase domain. The most common resistance mutation is the gatekeeper L1196M mutation, followed closely by the G1269A mutation. Other mutations occur at residues 1151, 1152, 1156, 1174, 1202, 1203, and 1206. The G1202R mutation is notable, as it confers resistance to crizotinib as well as to second-generation ALK inhibitors (alectinib, brigatinib, ceritinib, and ensartinib), but is sensitive to lorlatinib. (See 'Lorlatinib' below.)

●A second mechanism of crizotinib resistance is amplification of the ALK fusion gene. This can occur alone or in combination with a secondary resistance mutation.

●Activation of alternative or bypass signaling pathways has also been shown to mediate crizotinib resistance. These include abnormalities in EGFR, MET, KIT, and IGF-1 receptor pathways, and suggest the potential need for combination therapies to overcome resistance.

TREATMENT AFTER PROGRESSION ON SECOND-GENERATION ALK TKIS

Lorlatinib — Lorlatinib is a third-generation ALK tyrosine kinase inhibitor (TKI), which has been granted US Food and Drug Administration (FDA) approval for the treatment of patients with metastatic ALK-positive NSCLC [44]. Lorlatinib overcomes most acquired resistance mutations within the ALK kinase domain. Lorlatinib was also designed to be highly central nervous system penetrant. (See 'Lorlatinib' above.)

Because lorlatinib has activity against most of the known ALK inhibitor resistance mutations, including G1202R (one of the more common and more recalcitrant ALK kinase domain mutations), it is our preferred agent in the setting of acquired resistance to alectinib [38,65,66]. This mutation confers resistance to other next-generation ALK inhibitors, including ceritinib, alectinib, and likely brigatinib [67]. However, it is worth noting that the FDA approval of lorlatinib was not biomarker based, meaning that the presence of an acquired ALK kinase domain resistance mutation is not required to prescribe lorlatinib. Lorlatinib has activity both in patient with and without detected ALK kinase domain resistance mutations [68].

The FDA approval for lorlatinib is based on the results of a phase II study (B7461001) that included a subgroup of 198 patients with ALK-positive metastatic NSCLC previously treated with ≥1 ALK inhibitor [69]. The overall response rate with lorlatinib among these patients was 47 percent, with a complete response rate of 2 percent and a partial response rate of 45 percent. Efficacy according to prior treatment was as follows, at a median follow-up of approximately seven months: post-crizotinib, objective response rate (ORR) 73 percent, median progression-free survival (PFS) 11.1 months, and median duration of response not reached.

After one or more second-generation ALK inhibitors, the ORR was 40 percent, median PFS was 6.9 months, and median duration of response was 7.1 months. Similar findings were observed at a median follow-up of >30 months, and median overall survival (OS) in this subset was 21 months [70].

The most frequent adverse effects were hypercholesterolemia (81 percent), hypertriglyceridemia (61 percent), edema (43 percent), and peripheral neuropathy (30 percent) [69]. Serious treatment-related adverse effects occurred in 7 percent of patients, the most frequent being cognitive deficits in 1 percent. Among patients who had experienced progression on a prior second-generation ALK TKI, ORR to lorlatinib was higher among those with ALK mutations (in addition to the ALK rearrangement) than those without, suggesting that tumor genotyping for ALK mutations after failure of a second-generation ALK inhibitor may identify patients who are more likely to benefit from lorlatinib [68].

The intracranial efficacy of lorlatinib is discussed elsewhere. (See "Brain metastases in non-small cell lung cancer", section on 'Alectinib-resistant setting'.)

Although lorlatinib has not been compared with chemotherapy in the setting of alectinib-resistant disease, we prefer use of a next-generation ALK inhibitor, if available, given the poor survival outcomes that have been observed with chemotherapy in the setting of resistance to earlier-generation TKIs [54].

Alternative targeted therapies — Ceritinib and brigatinib have both shown activity, in small observational studies, among patients with progression on alectinib.

●A single-arm phase II study and retrospective data support limited clinical activity of brigatinib in the alectinib-refractory setting [71,72]. In a Japanese study including 47 patients with ALK-positive NSCLC refractory to alectinib, the ORR to brigatinib was 34 percent, and median PFS was 7.3 months. In a retrospective study of 22 patients with ALK-positive NSCLC and progression on alectinib, treatment with brigatinib was associated with objective responses in 17 percent and stable disease in 50 percent, and the median PFS was 4.4 months [71].

Later-line therapy: Chemotherapy, with or without immunotherapy — Upon progression on targeted agents, we now offer the combination of chemotherapy with immunotherapy based on subset analysis of IMpower 150, though we recognize that data regarding the efficacy of immunotherapy are limited and conflicting for those with activating driver mutations. Therefore, chemotherapy alone remains an acceptable alternative upon progression on available targeted agents and has shown modest activity in such patients [73].

In the IMpower 150 trial, the addition of atezolizumab to the combination of bevacizumab, carboplatin, and paclitaxel improved PFS and OS among 800 patients with advanced NSCLC [74]. The PFS benefit was also seen in the subset of 111 patients with epidermal growth factor receptor (EGFR) mutations or ALK rearrangements, all of whom had progressed on a prior targeted agent (median PFS, 9.7 versus 6.1 months; hazard ratio [HR] 0.59, 95% CI 0.37-0.94). These patients also experienced a trend toward improved OS (not estimable versus 17.5 months; HR, 0.54), though the difference did not reach statistical significance [75]. In the overall study population, serious adverse events occurred in 25 percent of those who also received atezolizumab versus 19 percent of those receiving standard therapy only.

While these data support carboplatin, paclitaxel, bevacizumab, and atezolizumab, some UpToDate experts offer carboplatin, pemetrexed, and pembrolizumab for patients with ALK-positive NSCLC after progression on available TKIs. This regimen may be more tolerable than the four-drug regimen, although it is recognized that trials evaluating carboplatin and pemetrexed with pembrolizumab excluded patients with ALK-positive NSCLC. Further details of these studies are described elsewhere. (See "Initial management of advanced non-small cell lung cancer lacking a driver mutation", section on 'In nonsquamous cancers'.)

In contrast to IMpower 150, the addition of atezolizumab to initial chemotherapy (nabpaclitaxel and carboplatin) did not improve PFS (HR 0.75, 95% CI 0.36-1.54) or OS (HR 0.98, 95% CI 0.41-2.31) in the subset of 44 patients with EGFR- or ALK-positive NSCLC in IMpower 130 [76]. We await further data for clarification on the role of immunotherapy for those with activating driver mutations upon progression on available targeted agents.

SPECIAL CONSIDERATIONS REGARDING TREATMENT

Brain metastases — For asymptomatic and symptomatic patients, second- or third- generation ALK tyrosine kinase inhibitors (TKIs) should be used, as data exist for each of these drugs regarding brain penetration and central nervous system efficacy. Most patients with brain metastases (either TKI naϊve or on crizotinib) will respond to these agents, and surgery and or/radiation therapy may be deferred, thereby potentially reducing patient morbidity associated with these local treatments. However, in the case of severe mass effect or impending herniation, we proceed with surgery as the initial treatment. Full discussion of these issues is found elsewhere. (See "Brain metastases in non-small cell lung cancer", section on 'ALK translocations'.)

Considerations during the COVID-19 pandemic — The COVID-19 pandemic has increased the complexity of cancer care. Important issues include balancing the risk from treatment delay versus harm from COVID-19, ways to minimize negative impacts of social distancing during care delivery, and appropriately and fairly allocating limited health care resources. These and recommendations for cancer care during active phases of the COVID-19 pandemic are discussed separately. (See "COVID-19: Considerations in patients with cancer".)

MANAGEMENT OF TOXICITIES ASSOCIATED WITH ALK INHIBITORS — Treatment with the ALK inhibitors is generally well tolerated. However, there are a number of significant toxicities that may require dose modification or treatment discontinuation.

●Gastrointestinal toxicity – Nausea, vomiting, and diarrhea are common with crizotinib and ceritinib, but are more severe with ceritinib. With crizotinib and alectinib, less than 1 percent of these side effects were severe. However, more than one-half of patients treated with ceritinib required dose modification, and 9 percent required treatment discontinuation. (See "Chemotherapy-associated diarrhea, constipation and intestinal perforation: pathogenesis, risk factors, and clinical presentation", section on 'ALK inhibitors'.)

Liver function test abnormalities can be seen with all of the ALK inhibitors. Transaminase elevation is commonly seen with crizotinib, ceritinib (particularly with ceritinib 750 mg, rather than 450 mg), and brigatinib. Elevated blood bilirubin can be seen with alectinib. Liver function tests should be checked every two weeks for the first eight weeks of treatment. Hepatotoxicity can develop and progress, and may require dose modification or discontinuation. (See "Chemotherapy hepatotoxicity and dose modification in patients with liver disease: Molecularly targeted agents", section on 'ALK inhibitors'.)

Constipation has been observed in over one-third of patients receiving alectinib. (See 'Preferred options' above.)

●Pulmonary toxicity – Severe, life-threatening, or fatal treatment-related pneumonitis has been reported in 1 to 4 percent of patients treated with either crizotinib or ceritinib. Alectinib is associated with an 0.4 percent incidence of interstitial lung disease (ILD)/pneumonitis, whereas brigatinib has a higher incidence (with reported frequency ranging between 3 and 9.1 percent) [29]. Of note, brigatinib has an unusual early pulmonary toxicity reaction (respiratory symptoms after one to two doses, accompanied by the development of ground-glass opacities on imaging) [40]. This toxicity is ameliorated by step-up dosing, as described above. (See 'Brigatinib' above.)

For any patient with new or worsening respiratory symptoms, brigatinib should be held, with prompt evaluation for ILD/pneumonitis or other causes of respiratory symptoms (eg, pulmonary embolism, tumor progression, and infectious pneumonia). For grade 1 or 2 ILD/pneumonitis, brigatinib may either be resumed with dose reduction after recovery to baseline, or permanently discontinued. For grade 3 or 4 ILD/pneumonitis or recurrence of grade 1 or 2 ILD/pneumonitis, brigatinib should be discontinued. (See "Pulmonary toxicity associated with antineoplastic therapy: Molecularly targeted agents", section on 'ALK inhibitors'.)

●Cardiac toxicity – Cardiac toxicity is observed with crizotinib, brigatinib, and ceritinib. Our approach is to perform an electrocardiogram (EKG) at baseline in all patients treated with these agents. Subsequently, we check an EKG if the patient develops bradycardia (whether symptomatic or not) or if the patient is started on a drug that is known to cause QTc prolongation. (See "Cardiotoxicity of cancer chemotherapy agents other than anthracyclines, HER2-targeted agents, and fluoropyrimidines", section on 'Crizotinib and ceritinib' and "Cardiotoxicity of cancer chemotherapy agents other than anthracyclines, HER2-targeted agents, and fluoropyrimidines", section on 'Brigatinib'.)

•Sinus bradycardia is relatively frequent in patients treated with crizotinib and may be severe in some cases [77,78]. However, patients with sinus bradycardia were asymptomatic, and the bradycardia was not explained by other comorbidity or medications. Caution should be used in the concomitant administration of beta blockers in patients treated with crizotinib. Ceritinib, alectinib, and brigatinib can also cause bradycardia. (See "Cardiotoxicity of cancer chemotherapy agents other than anthracyclines, HER2-targeted agents, and fluoropyrimidines", section on 'Crizotinib and ceritinib'.)

•QTc interval prolongation has been observed with crizotinib, and crizotinib should be avoided in patients with congenital long QT syndrome. Treatment should be temporarily discontinued if severe QTc prolongation develops and permanently discontinued if it recurs or is accompanied by an arrhythmia, heart failure, hypotension, shock, syncope, or torsade de pointes.

•Hypertension – Hypertension can be observed with brigatinib. In patients receiving the standard dosing regimen of brigatinib (90 mg run in, then 180 mg daily), 21 percent of patients developed hypertension, and 5.9 percent of patients had grade 3 hypertension. Blood pressure should be controlled prior to treating with brigatinib and should be monitored at least monthly. As brigatinib can cause bradycardia, antihypertensives that cause bradycardia should be used with caution.

●Musculoskeletal toxicity – Alectinib can cause myalgias, including muscle pain, tenderness, or weakness, particularly in the first month of treatment. These usually subside as treatment continues, but if the muscle symptoms are severe, alectinib should be held and may require dose reduction. The muscle symptoms may or may not be accompanied by creatine kinase elevation. Blood levels of creatine kinase should be tested every two weeks for the first month of treatment and in patients reporting muscle symptoms. Brigatinib can also cause myalgias and creatine kinase elevation; creatine phosphokinase levels should be monitored in patients with symptoms.

●Metabolic toxicity – Crizotinib, lorlatinib, and ceritinib are metabolized by cytochrome P450 3A4 (CYP3A4). Thus, caution should be used when these agents are given concomitantly with CYP3A4 inhibitors, and care is required when these agents are coadministered with other agents that are predominantly metabolized by this system (table 1). Lorlatinib has also been associated with increased hepatotoxicity when administered with CYP3A4 inducers; therefore, coadministration with strong CYP3A4 inducers is contraindicated. (See "Chemotherapy hepatotoxicity and dose modification in patients with liver disease: Molecularly targeted agents", section on 'ALK inhibitors'.)

●Hypercholesterolemia – Hypercholesterolemia is the most frequent toxicity associated with lorlatinib. In the phase II trial of lorlatinib, hypercholesterolemia occurred in 81 percent and hypertriglyceridemia occurred in 60 percent of patients, and each were severe in 16 percent. Recommended management includes treatment with beta-hydroxy beta-methylglutaryl-CoA reductase inhibitors, with rosuvastatin or pravastatin being preferred agents. Hypertriglyceridemia should be managed with fibrates, not gemfibrozil [79]. (See 'Lorlatinib' above.)

●Visual toxicity – Visual disturbances were seen in 60 to 65 percent of patients treated with crizotinib in phase II studies [22]. The main complaints are trailing lights, flashes, and brief image persistence, primarily associated with the transition from dark to light. Uncommon visual manifestations included photophobia, decreased visual acuity, and blurred vision. Visual symptoms have been reported with ceritinib, but their overall incidence is less than 10 percent.

These findings are not typically accompanied by changes on ophthalmologic examinations, and symptoms often improve over time; treatment cessation is usually not needed. However, the United States Prescribing Information recommends discontinuation of crizotinib in patients with new onset of severe visual loss [22]. (See "Ocular side effects of systemically administered chemotherapy" and "Ocular side effects of systemically administered chemotherapy", section on 'Anaplastic lymphoma kinase inhibitors'.)

●Neurologic toxicity – Central nervous system (CNS) effects of any cause have been observed in approximately 35 to 40 percent of patients on lorlatinib [69,80]. CNS effects included changes in cognitive function (eg, forgetfulness or difficulties with multitasking), mood disturbance (eg, mood lability and/or irritability), and speech disturbance (eg, word-finding difficulties or slowing of speech). These effects were generally mild, transient, and fully reversible with holding treatment. Notably, these side effects were also dose dependent. Therefore, for moderate or severe toxicity, our approach is to hold lorlatinib treatment until improvement, and resume treatment at a lower dose [79].

●Endocrine toxicity – Endocrine complications have been reported with crizotinib.

•With crizotinib, rapid depression of serum testosterone levels has been observed [81,82]. In a series of 32 patients treated with crizotinib, the mean serum testosterone level was below the lower limit of normal in 27 of 32 men (84 percent) compared with 6 of 19 (32 percent) among those not receiving crizotinib [82]. Levels of follicle-stimulating hormone and luteinizing hormone were also low, suggesting a centrally mediated mechanism. A large component of the lowered total testosterone may be due to significant depression of sex hormone-binding globulin. Male patients who report symptoms of hypogonadism can be referred to endocrinology to discuss potential testosterone replacement. (See "Clinical features and diagnosis of male hypogonadism".)

●Other toxicities

•Renal cysts – Renal cysts have been reported in 4 percent of patients treated with crizotinib [22]. The cysts may be positive on positron emission tomography and can create confusion with renal metastases [83]. These cysts may be complex. Although the natural history of crizotinib-associated renal cysts is not fully known, one report found that complex renal cysts regress when treatment with crizotinib is discontinued [84]. (See "Simple and complex kidney cysts in adults".)

•Hypersensitivity reactions – Hypersensitivity reactions have been described in two patients treated with crizotinib, manifested by a generalized rash developing hours after the administration of oral dosing [85]. Desensitization was carried out using escalating doses of crizotinib administered every 15 minutes over a three-hour period, preceded by one hour with single doses of loratadine (10 mg), cetirizine (10 mg), and fexofenadine (180 mg). Following desensitization, there were no further symptoms suggesting a hypersensitivity reaction.

•Photosensitivity reactions – Brigatinib has been associated with photosensitivity reactions. For example, in ALTA1L, photosensitivity occurred in 3.7 percent of patients receiving brigatinib, with 0.7 percent being grade 3 or 4 [42]. Patients should be advised to limit sun exposure while taking brigatinib, to wear a hat and sun-protective clothing while outdoors, and to use broad-spectrum sunscreen. Depending on the severity of the reaction, brigatinib may need to held, dose reduced, or discontinued.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Diagnosis and management of lung cancer".)

SUMMARY AND RECOMMENDATIONS

●Introduction

•Anaplastic lymphoma kinase (ALK)-positive non-small cell lung cancer (NSCLC) is a molecular subset of NSCLC with distinct clinical and pathologic features. ALK rearrangements are often but not exclusively found in lung tumors from relatively young, never or light smokers with adenocarcinoma. However, all patients with advanced nonsquamous NSCLC should be tested for ALK rearrangements. (See 'Clinicopathologic features' above.)

•In the United States, next-generation sequencing assays are strongly favored, as they can simultaneously assess multiple actionable biomarkers concurrently. Other approved methods of affirming ALK positivity include fluorescence in situ hybridization using Vysis Probes and immunohistochemistry using the Ventana ALK (D5F3) companion diagnostic assay. In Europe, immunohistochemistry is widely used to detect ALK rearrangement. (See 'Diagnosis' above.)

●Initial treatment – For patients with advanced or metastatic ALK-positive NSCLC, we recommend initial treatment with an ALK inhibitor rather than chemotherapy (Grade 1A). (See 'Rationale for ALK inhibitors' above.)

•In some cases, chemotherapy is required before the results of molecular testing are available. If the tumor is found to contain the ALK fusion oncogene, we suggest switching to an ALK inhibitor rather than continuing the present line of treatment (Grade 2C). (See 'Rationale for ALK inhibitors' above.)

•For those with newly diagnosed, ALK-positive NSCLC, we recommend a second- or third-generation ALK inhibitor, if available, rather than crizotinib (Grade 1B). In choosing between second-generation inhibitors, we typically suggest alectinib (Grade 2C), given the longer-term data with this agent compared with others and the toxicity profile. However, direct comparisons between second-generation inhibitors have not been performed, and we recognize brigatinib and lorlatinib are other acceptable first-line options. Ceritinib is approved in the front-line setting but is not preferred. (See 'First-line treatment' above.)

•Therapy should be continued until progression or prohibitive toxicity. In carefully selected patients with an isolated site of recurrence amenable to local therapy, or those with extremely mild and asymptomatic progression, the chosen ALK inhibitor may be continued after initial progression. (See 'Duration of treatment' above.)

●Subsequent-line treatment

•Data regarding management after progression on second-generation ALK inhibitors are evolving. For patients with ALK-positive NSCLC and progression on alectinib, we suggest lorlatinib over chemotherapy and other ALK inhibitors (Grade 2C). However, agents such as brigatinib and ceritinib are also active against some ALK resistance mutations, and are alternatives in these settings. (See 'Treatment after progression on second-generation ALK TKIs' above.)

•For patients with progression on available targeted agents and who remain candidates for further therapy, we opt for platinum-based chemotherapy alone, or in combination with bevacizumab and/or programmed cell death ligand 1 (PD-L1) inhibitors. The anti-PD-L1 antibody atezolizumab added to bevacizumab and platinum-based doublet chemotherapy has improved progression-free survival in nonsquamous NSCLC, including among those with ALK-positive NSCLC, though data are mixed. Some UpToDate contributors consider the combination of carboplatin, pemetrexed, and pembrolizumab upon progression on available tyrosine kinase inhibitors, recognizing, however, that this regimen has not been evaluated in patients with ALK-positive NSCLC. (See "Subsequent line therapy in non-small cell lung cancer lacking a driver mutation" and "Initial management of advanced non-small cell lung cancer lacking a driver mutation", section on 'In nonsquamous cancers'.)

●Special treatment considerations

•For ALK-positive patients who develop resistance to or are unable to tolerate crizotinib, we recommend treatment with one of the second-generation ALK inhibitors: brigatinib or alectinib (Grade 1B). Our preference in this setting is alectinib, given the central nervous system and systemic efficacy as well as tolerability, though, in the absence of head-to-head comparisons, any of these agents is an appropriate option. (See 'Ceritinib' above and 'Preferred options' above.)

•For most patients with brain metastases from NSCLC associated with an ALK rearrangement, we treat with a brain-penetrable ALK inhibitor. Surgery and/or radiation therapy may be pursued if urgent decompression is needed. (See "Overview of the treatment of brain metastases" and "Brain metastases in non-small cell lung cancer", section on 'ALK translocations'.)

●Toxicities of ALK inhibitors – ALK inhibitors are generally well tolerated, though associations with pneumonitis, liver function test abnormalities, bradycardia, and visual disturbances have been noted. Several significant toxicities may require dose modification or treatment discontinuation. (See 'Management of toxicities associated with ALK inhibitors' above.)

ACKNOWLEDGMENT — The editorial staff at UpToDate acknowledge Alice T Shaw, MD, PhD, who contributed to an earlier version of this topic review.