At present, for patients with advanced ALK-positive NSCLC who are resistant to the first-generation ALK inhibitor crizotinib, the second-generation ALK inhibitors ceritinib, alectinib, and brigatinib or the third-generation ALK inhibitor lorlatinib are recommended (9). Central nervous system metastasis was often inevitable for patients with crizotinib resistance, and the prognosis of which was relatively poor (10). In this study, 81.82% of patients without baseline brain metastases had developed brain metastases during treatment of crizotinib, indicating that central nervous system progression is a major challenge faced in the application of the first-generation ALK inhibitor. In previous studies (6, 11, 12), the incidence of central nervous system metastasis was 15%, 6.8% and 1% respectively in ceritinib, alectinib and brigatinib in treatment-naïve patients with ALK-positive NSCLC without baseline brain metastasis, which were relatively low, showing better control in intracranial lesions. Then, do second-generation ALK inhibitors also show favorite efficacy in crizotinib resistant patients with brain metastases? In this study, for 12 patients with brain metastases, ceritinib was used in posterior line, and the response for intracranial lesions was documented CR for 1 case, PR for 7 cases, and SD for 1 case, ORR was 88.89%, and DCR was 100%, indicating that the second-line therapy with ceritinib for patients with ALK-positive brain metastases had a good efficacy. Brain radiotherapy is also one of the most effective means for treatment of brain metastases (13), and whether radiotherapy will affect the efficacy of ALK inhibitors is also what we concerned. In this study, a total of 3 patients had received brain radiotherapy previously, all of whom had documented partial remission, suggesting that for those who had received brain radiotherapy, ceritinib may also be a good choice, i.e., the radiotherapy may not affect the efficacy of ceritinib. In this study, for the patients Who had received at least one or more regimens previously, the median PFS was 7.4 months, and the median OS 25.4 months. The PFS is much longer than that for the second-line therapy in the ASCEND-5 study (5.4 months) (6), and the reason may be that the 450mg daily dose of ceritinib taken with meal improved the safety and tolerability of the patients, which is yet to be verified by large sample randomized controlled trials.
In terms of dose selection, based on the ASCEND-8 study (8), ceritinib, 450mg taken with meal reduced the frequency and severity of toxicities, and increase the patients’ tolerance without compromising efficacy, and so, the ceritinib 450mg taken with meal was adopted. In our study, common adverse events were elevation of liver enzyme, diarrhea, and nausea, which was consistent with the previous report (14). Although the incidence of adverse events was high (91.67%), all of the adverse events were of grade 1 or 2, i.e., no grade 3 or 4 adverse events occurred, and no dose adjustment or drug withdrawal occurred due to adverse reactions, showing good safety and tolerability. Therefore, ceritinib 450mg taken with meals is a good choice worthy of clinical application.
At present, there is still no clinical reports about the combination of ALK inhibitor with antiangiogenic drug. Bevacizumab is a monoclonal antibody against vascular endothelial growth factor (VEGF), and it was demonstrated by previous phase II trials that the combination of bevacizumab with erlotinib could prolong the PFS of patients with advanced EGFR-mutation NSCLC (15, 16), And the addition of bevacizumab to EGFR-TKIs also showed efficacy in progression-free survival of patients with EGFR-TKIs resistance (17). For patients with brain metastases or brain edema, bevacizumab also showed a good efficacy, and the response rate in cerebral edema was 84.74% (18). In this study, we tried to use ceritinib in combination with bevacizumab in the third-line therapy for 1 patient, unfortunately, the patient died only 1.8 months after medication since the patient’s had been in poor conditions after multiple lines of treatment, unable to assess the efficacy. Of course, for the treatment in combination with bevacizumab, we need to pay more attention to relevant toxic and side effects, especially hypertension, proteinuria, etc (19, 20). Whether the combination of ALK inhibitors with antiangiogenic drugs, e.g. bevacizumab, anlotinib, could bring survival benefits for patients is still to be explored by more clinical studies in the future.
In non-small cell lung cancer, the most common site for ALK fusion is the fusion of echinoderm microtubule-associated protein-like 4 (EML4) with ALK (21), other rare fusion sites include the fusion of ALK with TFG, KLC1, KIF5B, STRN, TNFAIP3, TNIP2, etc.(22–24). There were also case reports concerning efficacy of ALK inhibitors on rare ALK fusion previously (22, 25). The two cases of rare ALK fusion found in this study were PKNOX-ALK fusion and IGR-ALK fusion cases, and both cases had been treated by crizotinib effectively, with PFS 11.5 months and 5 months respectively. The ceritinib treatment after crizotinib resistance was still effective, with intracranial lesions significantly reduced. The two cases received ensartinib and alectinib respectively as posterior line therapy, both of which showed efficacy. With the wide application of next-generation sequencing, more and more rare ALK mutations have been found in clinical practice. For both cases of rare ALK fusion in this study, the first-generation ALK inhibitor and the two second-generation ALK inhibitors were used effectively, which has brought us a great deal of confidence in these two kinds of rare ALK fusion, but can all of the patients with rare ALK fusion benefit from the first generation to the third generation of ALK inhibitors? There are only some individual cases for now, and evidence from clinical trials is to be supplemented.
It is well known that all targeted therapies may be faced with drug resistance at last. The resistance mechanism of the first-generation ALK inhibitor crizotinib mainly involves the following aspects: secondary mutation of ALK, ALK bypass activation, and NSCLC transformation. Secondary mutation of ALK includes L1196M, C1156Y, G1269A, S1206Y, V1180L, G1202R, I1171T, D1203N, 1151Tins, F1174C/L, L1198P, L1152R/P and E1210K (26). I1171 mutation indicates resistance to crizotinib and alectinib (27), and G1202R or I1171N mutation often occurs after alectinib resistance (28), while I1171N site mutation was sensitive to ceritinib (29). ALK secondary mutation was found in about half of patients with ceritinib mutation, of which, G1202R (21%) andF1174C/L (16.7%) are most common (29). Furthermore, some patients had more than two kinds of ALK secondary mutation. The T1151R mutation indicates sensitivity to brigatinib (30). G1202R mutation often occurs after brigatinib resistance as well (31) and it suggests resistance to both the first generation and second-generation ALK inhibitors (32). The third-generation ALK inhibitor is sensitive to not only F1174, C1156Y, G1269A, L1196M, D1203N mutations, but also G1202R mutation, therefore, for such patients, the third-generation ALK inhibitor lorlatinib may be recommended (33). The mechanism of lorlatinib resistance is more complex and variable, including L1196M/D1203N, F1174L/G1202R, and C1156Y/G1269A mutations, etc. Patients with a single mutation of ALK-L1256F may be sensitive to ceritinib. Generally speaking, the ALK secondary mutation rate is approximately 20-30%, while the response rate of the second-generation ALK inhibitors was (48-71%) (28), indicating that the second-generation ALK TKIs may be also effective in crizotinib-resistant patients without ALK secondary mutation. In this study, 4 patients underwent genetic testing after first-generation ALK inhibitor resistance, and no ALK secondary mutation was found among them, but the second-generation ALK inhibitor ceritinib was still effective. Therefore, for cases resistant to the first-generation ALK inhibitor, the treatment with the second-generation ALK inhibitors should be considered even in irst-generation ALK inhibitor resistant patients without genetic testing. It could also maybe the sensitivity of the NGS detection method using blood samples is not as good as using tissue samples. If genetic testing can be performed again on new lesion’s biopsy sample to determine the presence of ALK secondary mutation and specific mutation site, the subsequent treatment may be guided more accurately. Besides ALK second mutation, other kinases mutations and bypass signal abnormal activations (e.g., TP53 mutation, MET mutation, BRAF mutation and MAPK, PI3K/Akt pathway and HER pathway activation) are important mechanisms of resistance for ALK positive patients (26) (34), reflecting the diversity and complexity of resistance to ALK inhibitors, which has not been fully studied yet. (26) (26) However, it has been shown that the combination of proteasome inhibitors and alectinib is effective in patients with TP53 mutation after alectinib resistance (35). Some studies have found that a bypass mechanism of resistance caused by NF2 loss-of-function mutations is sensitive to mTOR inhibitor (33). Another mechanism of resistance is the small cell lung cancer transformation. For patients with dramatically elevated NSE or rapid progressive disease or cases with a mutation for dual deletion of RB1 and TP53 during treatment, a second pathological biopsy is recommended for detecting whether there is transformation of small cell lung cancer (32). For patients with rapidly progressive small cell lung cancer transformation, chemotherapy with etoposide combined with cisplatin could be considered, for patients with less progressive disease or local progression, chemotherapy in combination with original TKI targeted therapy may be appropriate, and local treatment may be supplemented if necessary (36).