Incidence of pneumonitis following the use of different anaplastic lymphoma kinase tyrosine kinase inhibitor regimens: An updated systematic review and meta‐analysis

Abstract Objectives Anaplastic lymphoma kinase tyrosine kinase inhibitors (ALK TKIs) have shown remarkable clinical activity in patients with non‐small‐cell lung cancer (NSCLC). However, pneumonitis is a serious side effect of ALK TKIs in NSCLC patients. In this meta‐analysis, we aimed to determine the incidence of ALK‐TKI‐associated pneumonitis. Materials and Methods We searched electronic databases to identify relevant studies published until August 2022. The incidence of pneumonitis was calculated using a fixed‐effects model when no substantial heterogeneity was observed. Otherwise, a random‐effects model was used. Subgroup analyses of different treatment groups were performed. Statistical analyses were conducted using STATA 17.0. Results Twenty‐six clinical trials involving 4752 patients were eligible for analysis. All‐grade pneumonitis incidence was 2.92% (95% confidence interval [CI]: 1.79%–4.27%), high‐grade (Grade 3–4) pneumonitis incidence was 1.42% (95% CI: 0.84%–2.12%) and Grade 5 pneumonitis incidence was 0.09% (95% CI: 0.00%–0.28%). The subgroup analysis showed that brigatinib was associated with the highest incidence of both all‐grade and high‐grade pneumonitis (7.09% and 3.06%, respectively). ALK TKI treatment after chemotherapy was associated with a higher incidence of all‐grade and high‐grade pneumonitis than first‐line ALK TKI treatment (7.73% vs. 2.26% and 3.64% vs. 1.26%, respectively). Cohorts from Japanese trials had a higher incidence of all‐grade and high‐grade pneumonitis. Conclusion Our study provides precise data on the incidence of pneumonitis in patients receiving treatment with ALK TKIs. Overall, ALK TKIs have tolerable pulmonary toxicity. Early pneumonitis identification and treatment are required to prevent further deterioration in patients receiving treatment with brigatinib and in those who received prior chemotherapy, particularly in the Japanese population.


| Data sources and searches
Searches were conducted in electronic databases, including PubMed, Web of Science, Embase and Cochrane Library, for relevant articles published until August 2022. The following search term, which combined synonyms using the keywords, "crizotinib," "brigatinib," "lorlatinib," "ceritinib," "alectinib," and "lung cancer," was used: (crizotinib OR brigatinib OR lorlatinib OR ceritinib OR alectinib) AND (lung cancer). These five ALK TKIs were chosen for the study because at the time of the study, the FDA had approved these drugs for the treatment of NSCLC with ALK rearrangement.

| Study selection and data extraction
The inclusion criteria were as follows: (1) studies involving patients with advanced NSCLC who were receiving approved doses of ALK TKIs in clinical trials and (2) studies providing detailed information on ALK-TKI-associated pneumonitis. Conference abstracts, brief reports, case reports, reviews and non-English articles were excluded. Studies involving neoadjuvant therapy or the use of ALK-TKIs in conjunction with other anticancer treatment strategies were excluded. In case different studies involved the same population, the most recent and relevant article was chosen for our meta-analysis.
Two investigators independently assessed all the trials and collected the following information from the identified articles: initial author's name, publication year, region and treatment line. The treatment lines were divided as follows: (1) first-line ALK TKI therapy without any prior systemic therapy, (2) ALK TKI therapy following prior chemotherapy without any prior ALK TKI therapy and (3) ALK TKI therapy following prior chemotherapy or other ALK TKI therapy (previously treated group). The details of all-grade, high-grade (Grade 3-4) and Grade 5 pneumonitis-related deaths were collected. In addition, the number of patients with cough and dyspnoea was recorded.

| Quality assessment and statistical analysis
The Newcastle-Ottawa Scale (NOS) was used to evaluate the quality of the included studies, and studies with an NOS score ≥6 were considered to be of high quality. 19 We used a systematic analytical approach to compute the pooled pneumonitis rates in all the eligible studies. The weights of each study were determined using a random-effects model directly from the STATA 'metaprop' command. Before pooling, the variances were stabilised using the Freeman-Tukey double arcsine transformation method. 20 Heterogeneity was evaluated using the Q statistic and Higgins inconsistency index (I 2 ) test. A p-value of 0.05 was used to assess the test's statistical significance. An I 2 greater than 50% was considered to represent a large amount of heterogeneity, in which case the source of heterogeneity was explored. The potential causes of heterogeneity were investigated using subgroup analyses and metaregression. All statistical analyses were performed using Stata version 17.0.

| Incidence of high-grade pneumonitis
Across all the study cohorts, 71 of 4431 patients had highgrade pneumonitis. The incidence of Grade 3 and highergrade pneumonitis in the 24 cohorts is shown in Figure 3. The pooled incidence was 1.42% (95% CI: 0.84%-2.12%).

| Incidence of dyspnoea and cough
The overall incidence of all-grade dyspnoea was 18.6%. The incidence was similar among the different ALK TKI groups (ranging from 16.63% in the brigatinib group to 20.13% in the crizotinib and lorlatinib groups). However, the probability of all-grade cough differed markedly among the different ALK TKI groups (Supplementary Figure 1). The highest incidence was observed in the brigatinib group (36.03%), and the lowest, in the lorlatinib group (16.11%).
There was considerable variation in the proportion of patients with high-grade dyspnoea between the different F I G U R E 2 Incidence of all-grade pneumonitis during treatment with anaplastic lymphoma kinase tyrosine kinase inhibitors.

| Publication bias
Slight asymmetry was visible in the funnel plots for each grade of pneumonitis, suggesting possible publication bias. The asymmetry of the funnel plots was further evaluated using Egger's test. There was evidence of publication bias based on the results of Egger's test (all p < 0.05).

| DISCUSSION
The findings of this meta-analysis reveal an association between the occurrence of pneumonitis and use of ALK TKIs. In our study, 2.92% of patients receiving treatment with ALK TKIs developed all-grade pneumonitis, 1.42% developed high-grade pneumonitis and 0.09% developed Grade 5 pneumonitis. To our knowledge, this is the most thorough systematic review and meta-analysis of the incidence of pneumonitis following an FDA-approved ALK TKI regimen in patients with ALK rearrangement in advanced NSCLC.
Suh et al. reported an all-grade pneumonitis incidence of 2.14%, a high-grade pneumonitis incidence of 1.33% and a Grade 5 pneumonitis incidence of 0.22% 17 ; in comparison, we identified a slightly higher rate of allgrade pneumonitis. Our findings related to high-grade pneumonitis are consistent with those of previous metaanalyses. However, the proportion of patients with Grade 5 pneumonitis has decreased. One potential reason for this proportional redistribution may be the recent focus on pneumonitis as a rare but serious side effect of ALK TKIs and the increasingly prompt management of this condition.
Our study has revealed for the first time that the lowest rate of all-grade and high-grade pneumonitis is associated with lorlatinib compared with other ALK TKIs.
T A B L E 2 Results of multiple subgroup analysis for the incidence of all-grade pneumonitis.

Subgroup
No A single-arm meta-analysis showed that lorlatinib had the lowest discontinuation rate (3%) compared with that of alectinib (7%), brigatinib (7%), ceritinib (8%) and crizotinib (8%). 40 This finding illustrates the pulmonary safety of lorlatinib particularly well. Brigatinib, a secondgeneration ALK TKI, was associated with the highest rate of pneumonitis occurrence. The ALTA study reported on the safety of brigatinib for treating crizotinib-refractory ALK positive NSCLC. All deaths occurred in the groups receiving 90 mg quaque die (QD), 120 mg QD and 180 mg QD and they occurred within 7 days of starting treatment; further, dose interruptions were required in a larger number of patients in the higher dose group. Therefore, a regimen involving run-in dose escalation from a small to a regular dose was developed. It has been presumed that the occurrence of pneumonitis is related to the dose and usage of treatment and that dose adjustment can control the occurrence of pneumonitis. 15 However, pneumonitis occurred relatively late with some other ALK TKIs. Crizotinib-induced pneumonitis has two clinical features. The first type is a severe and irreversible form of pneumonitis that develops early (usually within a month after the initial treatment) and with more severe symptoms. The second type is a late-onset form that typically occurs around 3 months after starting crizotinib. 41,42 Median time to lorlatinib-associated pneumonitis was 73.5 days. 43 However, there are currently no published data regarding the average time to pneumonitis associated with alectinib and ceritinib. It is important not to overlook pneumonitis due to its late onset. Chemotherapy remains the key treatment option for cancer. Therefore, we have presented data on the incidence of pneumonitis in patients who received chemotherapy before ALK TKI administration. We found a high incidence (7.73%) of all-grade pneumonitis in patients who had received prior chemotherapy. In addition, prior chemotherapy increased the rates of high-grade and Grade 5 pneumonitis; however, the difference was not significant. Our results are in accordance with the finding that chemotherapeutic agents, such as paclitaxel, also cause lung injuries, with an incidence below 1%. 44,45 However, a similar phenomenon was not observed in the previously treated group. We speculate that the difference in the number of studies included in each group may have caused heterogeneity and bias. The results of the present study would be more convincing if a larger number of randomized controlled trials (RCTs) could have been included. The mechanism underlying superposed lung damage may be complex. Hence, before receiving treatment with ALK TKIs, patients with a history of chemotherapy should undergo stricter screening in order to decrease the risk of pneumonitis and prevent fatal pneumonitis.
Pneumonitis is thought to affect Japanese patients more frequently than other ethnic groups. This result is in accordance with that of earlier reports. 17,46 Pneumonitis is also one of the most common SAE associated with epidermal F I G U R E 3 Incidence of high-grade pneumonitis during treatment with anaplastic lymphoma kinase tyrosine kinase inhibitors.
T A B L E 3 Results of multiple subgroup analysis for the incidence of high-grade pneumonitis.

Subgroup
No growth factor receptor-TKIs, with a documented incidence of 1.6%-4.3% in Japanese populations and 0.3%-1.0% in non-Japanese populations. 47 Krebs von den Lungen-6 has been reported to serve as a sensitive serum marker and is now used clinically to detect pneumonitis in Japan. 48 Moreover, genetic variation in surfactant protein-D (SP-D) is a risk factor for pneumonitis development in the Japanese population. 49,50 The German cohort had significantly higher serum SP-D levels than did the Japanese cohort. 51 Further scientific research is needed to determine the causes of the difference in incidence between Japan and the rest of the world. The study from Japan did not find a higher probability of Grade 5 pneumonitis; therefore, we speculate that the high prevalence of Grade 1-4 pneumonitis in Japanese patients is also related to early review and follow-up chest computed tomography.
The mechanism underlying the development of ALKinduced pneumonitis is yet to be fully elucidated; however, alveolar epithelial injury is thought to play a role. 52 Pneumonitis associated with ALK TKIs can be considered as allergic pneumonitis. A low cluster of differentiation (CD)4/CD8 ratio was observed in a large number of lymphocytes (TCD4/TCD8 ratio of 1.4, 90% lymphocytes) in bronchoalveolar lavage fluid collected from patients, which again indicates a hypersensitivity reaction. 41,53 In addition, ALK TKI-induced pneumonitis may be associated with an antitumour immune response. Patients with pneumonitis exhibited a positive tumour response. On comparing these individuals to those without pneumonitis, it was found that their PFS was longer (19.9 months and 6.2 months, respectively). 41 In general, restarting ALK-targeted therapy in patients with ALK-induced pneumonitis requires careful consideration. Currently there is a lack of evidence in the guidelines regarding retreatment with ALK TKIs. However, if ALK TKI therapy is stopped due to pneumonitis, the options for follow-up treatment become extremely limited, which can have a negative impact on patient survival. Nevertheless, there have been many successful cases of ALK TKI retreatment. The safety and efficacy in these patients could be guaranteed even after experiencing Grade 3 or higher pneumonitis. 54,55 Alternative ALK TKIs are commonly used in practice, as different ALK TKIs may have different mechanisms for the development of pneumonitis. 56 However, in cases where follow-up options are limited, some patients have also attempted re-challenge with ALK TKIs. For example, a 76-year-old woman developed severe pneumonitis and respiratory failure a month after initiating alectinib treatment. The patient discontinued alectinib and received mechanical ventilation and methylprednisolone in the intensive care unit. After recovery, a low dose of alectinib was administered. Rechallenge therapy was successful and no pneumonitis occurred after 480 days. 57 It is important to note that there are also risks associated with restarting ALK TKI therapy. According to another report, a patient who recovered from brigatinib-induced interstitial lung disease relapsed with pneumonitis immediately after starting lorlatinib treatment. 58 Further research is needed to determine whether re-challenging with ALK TKIs carries an increased risk of recurrent pneumonitis. Nevertheless, caution and early prevention measures must be taken when considering this approach in clinical practice. The aetiology of fatal pneumonitis remains largely unknown.
However, several questions remain to be addressed. First, nearly half of the patients with pneumonitis were classified as having high-grade pneumonitis. Additional retrospective studies are required to elucidate the susceptible T A B L E 4 Results of multiple subgroup analysis for the incidence of Grade 5 pneumonitis.  59 Other trials, such as those of alectinib combined with cobimetinib (NCT03202940), brigatinib combined with platinum-based chemotherapy (NCT05200481) and brigatinib combined with bevacizumab (NCT04227028), are ongoing. There is a need to further explore the incidence of AEs associated with ALK TKIs in combination with other drug treatments.

Subgroup
There are no published data on whether the addition of ALK TKIs can induce high-grade pneumonitis in patients who received prior radiation therapy. Our study has some limitations. First, subjective assessments of pneumonitis were performed in the different medical centres and our analysis depended on the quality of the investigators' reports. Heterogeneity was observed among the studies included in this review. In addition, we performed a meta-analysis at the study level; consequently, patient-level variables were not available for this analysis. Therefore, we were unable to identify other potential risk factors that may be associated with the development of pneumonitis, including the role of sex, baseline lung function, or smoking history. Larger multicentre RCTs are required to assess the safety of various ALK TKIs in patients with NSCLC.

| CONCLUSIONS
Our study provides more precise data on the incidence of pneumonitis induced by various ALK TKIs. Overall, ALK TKIs had a good pulmonary toxicity profile. Although the incidence of pneumonitis is high, most patients experience mild symptoms and the mortality rate is low after aggressive prevention and treatment. It should be noted that extra caution is needed to prevent the development of pneumonitis in patients being treated with brigatinib, those with a history of prior chemotherapy and those of Japanese ethnicity.