Efficacy and safety of RET-TKI in advanced RET-rearranged non-small cell lung cancer in China: a real-world retrospective chart review

Abstract

Background

Selective RET inhibitors have been approved by the Chinese government for the treatment of RET-rearranged non-small cell lung cancer. This study aimed to illustrate the efficacy and safety of selective RET inhibitors in a real-world clinical context in China.

Methods

Patients with locally advanced or metastatic non-small cell lung cancer (NSCLC) harboring RET rearrangement and receiving RET tyrosine kinase inhibitors (RET-TKI) in the real world were enrolled in this retrospective study. Clinical data, including baseline clinicopathological information, efficacy parameters such as objective response rate (ORR) and progression-free survival (PFS), and adverse events (AEs), were collected from the electronic medical record system. The pattern of treatment failure of first-line RET-TKI was also described.

Results

Fifty-one patients were enrolled in this study. RET-TKI induced an ORR of 73.1% and a median PFS (mPFS) of 22.7 months (95%CI, 11.7–33.7) in the first-line setting. The ORR and mPFS were 58.3% and 17.7 months (95%CI, 9.1–26.2), 55.6% and 14.7 months (95%CI, 12.6–16.8) in the second-line and later-line settings, respectively. No significant difference was observed among different application lines with respect to the ORR (P = 0.534) or PFS (P = 0.795). In the first-line setting, RET-TKI significantly prolonged PFS compared to other regimens including chemotherapy-based regimens, multikinase inhibitors and other systemic regimens without chemotherapy (P < 0.05). Poor ECOG performance status was related to shorter PFS (P = 0.018). The most common AEs of grade 3 or worse were a decreased neutrophil count (11.4%) and anemia (11.4%). No new AEs or grade 5 AEs were observed. Brain metastasis was one of the most common patterns of treatment failure. In patients with baseline brain metastasis, the intracranial ORR was 50%, and the DCR was 100%.

Conclusions

RET-TKI had favorable efficacy and safety in real-world contexts in China and should be considered the preferred choice for first-line treatment in RET-rearranged NSCLC patients.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-024-13155-z.

Introduction

Lung cancer remains one of the most common and deadly malignant tumors in China causing enormous health threats and economic losses every year [1]. Targeted therapy has profoundly changed the treatment landscape of non-small cell lung cancer (NSCLC) in well-selected patients harboring driver gene alterations [2], including the rearrangements during transfection (RET) gene, which was first identified in NSCLC in 2012 [3]. The incidence rate of RET rearrangement in NSCLC ranges from 1 to 2% [4], and KIF5B has been identified as a predominant upstream partner gene, accounting for more than 70% of all fusion partners, followed by CCDC6 and NCOA4 [5].

The treatment strategy for advanced NSCLC with RET rearrangement was once limited to platinum-based chemotherapy before the development of specific tyrosine kinase inhibitors or immune checkpoint inhibitors [6]. A global multicenter study including 165 NSCLC patients with RET rearrangement demonstrated that the objective response rate (ORR) of first-line platinum-based chemotherapy was 49% and the median progression-free survival (PFS) was 6.4 months [5]. In addition, pemetrexed-based regimens were found to increase PFS and a trend for overall survival (OS) benefit compared with other chemotherapy regimens in a retrospective study [7]. Recently, immunotherapy has emerged as an important treatment for advanced NSCLC [8]. The role of immunotherapy in NSCLC patients with RET rearrangement remains controversial on the basis of several small retrospective studies, with median PFS (mPFS) ranging from 2.1 to 7.6 months and ORR ranging from 6.3% to 37.5% [9–11]. However, in consideration of the risk for immune-related adverse events (AEs) induced by the combination of immunotherapy and RET-targeted therapy, immunotherapy is not prioritized in the Chinese expert consensus for patients with RET-rearranged NSCLC [12].

Nonselective multikinase inhibitors (MKI) have low efficacy in patients with RET-rearranged NSCLC [5, 13, 14], probably due to unfavorable pharmacokinetic profiles in this population or heavily treated patients enrolled [5]. On the other hand, RET-specific inhibitors, such as pralsetinib (BLU-667) and selpercatinib (LOXO-292), have been proven to induce durable clinical responses [15, 16], and have been approved by the China National Medical Products Administration for first-line use in advanced NSCLC patients with RET rearrangement. However, data concerning the efficacy and safety of RET-specific inhibitors in a real-world context in China are limited.

Therefore, it is necessary to add up to the clinical efficacy of RET-specific inhibitors for RET-rearranged NSCLC in the real-world context to seek more evidence and provide more suggestions to local clinicians in China.

Materials and methods

Clinical data collection

This was a retrospective study in a real-world setting. Medical records of patients in the National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital were retrieved from the Hospital Information System. The inclusion criteria were as follows: (1) histologically diagnosed with stage III or IV NSCLC; (2) had a tumor gene test revealing RET rearrangement; and (3) had a history of RET-TKI use. The exclusion criteria were as follows: (1) history of radical surgery or radical radiotherapy; (2) diagnosis of multiple primary malignancies; and (3) missing key information concerning treatment, survival and gene tests. The clinical, genetic, and treatment information of eligible patients were independently collected by two investigators. A third senior investigator was consulted to resolve any inconsistencies between the first two investigators. The details of the patient selection flow chart were presented in Fig. 1.

Fig. 1.

Flow chart of patient disposition. NSCLC, non-small cell lung cancer; RET, the rearranged during transfection gene; EGFR, epidermal growth factor receptor

Gene testing, radiologic and safety evaluation

The test methods included reverse transcriptase-polymerase chain reaction (PCR) and next-generation sequencing (NGS). Specifically, a total of 45 patients were tested with NGS, 1 patient was tested with both NGS and PCR and the testing method of the other 5 patients was unknown. Different NGS panels were used to identify the RET rearrangement, including a 520-gene panel (OncoScreen Plus®, Burning Rock Biotech Limited) used in 23 patients, a 769-gene panel (Genecast Biotechnology Co., Ltd) and a 639-gene panel (Yunying Medicine) used in one patient respectively. The specific NGS panel of the other 26 patients were unknown due to the retrospective nature of the clinical data. Imaging examinations at baseline included measurable target lesions documented by computed tomography (CT) images of the chest and abdomen, brain magnetic resonance imaging (MRI), and whole bone scans. The blood routine and biochemical examinations were performed during the systemic therapy. Efficacy was generally evaluated by imaging examination every two treatment cycles for patients who were treated with chemotherapy and two months for patients after ending chemotherapy, two or three cycles for patients treated with ICIs, and two months for patients treated with targeted therapy. The response to therapies was assessed in local hospitals according to the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1. Complete response (CR), partial response (PR), stable disease (SD) or progressive disease (PD) were evaluated in this study. PFS was defined as the duration from the initiation of therapy to the date of disease progression or death. ORR was defined as the sum of the percentages of CRs and PRs. The disease control rate (DCR) was defined as the sum of the percentages of CR, PR and SD patients. The response evaluation was conducted by one investigator and re-checked by another investigator. Safety was evaluated according to the Common Terminology Criteria for Adverse Events (version 5.0).

Statistical analysis

Patient characteristics at baseline were analyzed via descriptive statistics. The categorical variables were presented as frequencies and percentages and the chi-square test or Fisher’s exact test was used for comparison. The normally distributed data were displayed as the means and standard deviations and were compared via t-test. The median and 95% CI for PFS were determined via the Kaplan–Meier method and assessed via the log-rank test or Cox proportional hazards. An α value of 0.05 was used as the examination standard. SPSS Statistics version 25.0 was used to conduct statistical analysis and analytic charts were created with R version 4.2.2.

Ethics approval and consent to participate

This study was conducted in accordance with the Declaration of Helsinki. The Ethics Committee of National Cancer Center/ Cancer Hospital/ Chinese Academy of Medical Sciences and Peking Union Medical College approved the study and waived the informed consent for its retrospective noninvasive nature (Approval Number: 18–070/1648).

Results

Baseline characteristics

A total of 51 patients were enrolled. The diagnostic time ranged from April 2016 to July 2023 and the last follow-up date was 3rd May 2024. Table 1 summarized the baseline characteristics of these patients. The average age of the patients was 56.3 years (Standard deviation, 10.8 years). Twenty-nine patients were female (56.9%). Thirty-nine patients had no history of smoking (76.5%). Most patients had an Eastern Cooperative Oncology Group performance status (ECOG PS) of 0 (n = 25, 49.0%) or 1 (n = 22, 43.1%). Adenocarcinoma accounted for 92.2% (n = 47) of all patients. Thirty-five patients were diagnosed with stage IV disease (68.6%) and 23.5% (n = 12) of patients had recurrent disease. The median number of distant metastatic organs was 2 (interquartile range, 1–3). Approximately 20% of patients had brain metastasis (n = 10), and four patients had liver metastasis (7.8%). KIF5B was the most common upstream fusion partner (n = 33, 64.7%), followed by CCDC6 (n = 7, 13.7%). The most common co-alteration was TP53 mutation, occurred in 30.2% of patients, followed with PTEN mutation (4.6%) and MDM2 amplification (4.6%). A detailed heatmap of patients’ baseline characteristics was shown in Fig. 2.

Table 1.

Baseline characteristics of patients in this cohort (N = 51)

Baseline Characteristics	Number (%)
Age (mean (SD)), years	56.31 (10.87)
Sex
Male	22 (43.1)
Female	29 (56.9)
Smoking History
Never	39 (76.5)
Ever or current	12 (23.5)
ECOG PS
0	25 (49.0)
1	22 (43.1)
≥ 2	4 (7.8)
Histology
Adenocarcinoma	47 (92.2)
Adenosquamous carcinoma	1 (2.0)
Squamous cell carcinoma	1 (2.0)
Combined large cell neuroendocrine carcinoma	2 (3.9)
Stage
IV	35 (68.6)
Relapse	12 (23.5)
III	4 (7.8)
Number of distant metastatic organs (median [IQR])	2 [1, 3]
Brain metastasis
Absent	41 (80.4)
Present	10 (19.6)
Liver metastasis
Absent	47 (92.2)
Present	4 (7.8)
Upstream Fusion Partners
KIF5B	33 (64.7)
CCDC6	7 (13.7)
NCOA4	2 (3.9)
Othersa	4 (7.8)
Unknown	5 (9.8)

IQR interquartile range, ECOG PS Eastern Cooperative Oncology Group Performance Status

^aOther fusion partners included ABI3BP, JMJD1C, SPECC1L, FXYD4, NCKAP5L, ZNF33A and MPP7

Fig. 2.

A detailed heatmap of baseline characteristics of enrolled patients. ECOG PS, Eastern Cooperative Oncology Group Performance Status; RET, the rearranged during transfection gene

Overview of treatment strategies and the clinical course

The clinical course and treatment strategies in chronological order were displayed through a swimmer plot in Fig. 3. The majority of patients have received multiple-line treatment. To simplify the illustration, regimens were classified into different groups according to drug type and clinical experience. The regimens containing RET-TKI with or without the addition of other therapies were grouped into the “RET-TKI” group. In this group, forty-four (86.2%) patients used pralsetinib, six patients (11.7%) patients were treated with selpercatinib and one patient (2%) attended a clinical trial and applied HS-10365 (a small molecular, oral potent, selective RET inhibitor by Jiangsu Hansoh Pharmaceutical Co., Ltd.). The regimens containing chemotherapy were grouped to form the “chemo-based” group, which was further divided into the “chemo-alone” group, in which regimens consisted of only chemotherapy, and the “chemo-combined” group, in which regimens consisted of chemotherapy and other treatment methods such as immunotherapy and anti-angiogenesis therapy. The group consisting of regimens free of chemotherapy was named “chemo-free”, including immunotherapy, antiangiogenic therapy and the combination of immunotherapy and antiangiogenic therapy. Notably, MKI was not included in the “chemo-free” group and was classified into an independent group named “MKI”. The grouping methods continued to be used in the following data illustration and analysis. A visualization of the grouping method was shown in Supplemental Fig. 1.

Fig. 3.

A swimmer plot of the clinical course and treatment strategies in chronological order. RET-TKI, the tyrosine kinase inhibitor for rearranged during transfection gene; MKI, multikinase inhibitors

In the first-line setting, the RET-TKI was the most commonly applied treatment, accounting for 57% (29/51) of all therapeutic methods. Chemo-based therapy was subsequently used by 35% (18/51) of the patients, in which 12 (23.5%) patients received “chemo-combined” therapy and 6 (11.8%) patients received “chemo-alone” therapy. “Chemo-free” and MKI were used in only 4% of patients in the first-line setting. Detailed information on the therapeutic regimens was listed in Supplemental Table 1.

Efficacy of RET-TKI

Results of different application lines

In this cohort, 29 (57%) patients used RET-TKI as first-line treatment, 12 (23.5%) patients used RET-TKI in a second-line setting and 10 (19.5%) patients used RET-TKI in third- or later-line settings. In the first-line setting, 26 patients were available for efficacy evaluation. The ORR was 73.1% (19/26), the DCR was 96.2% (25/26) and the mPFS was 22.7 months (95%CI, 11.7–33.7). In the second-line setting, the ORR was 58.3% (7/12), the DCR was 100% (12/12) and the mPFS was 17.7 months (95%CI, 9.1–26.2). In the later-line setting, efficacy was evaluable in 9 patients. The ORR was 55.6% (5/9), the DCR was 100% (9/9) and the mPFS was 14.7 months (95%CI, 12.6–16.8). The ORR showed no significant difference among different application lines (P = 0.534), nor was the mPFS (P = 0.795) (Fig. 4A, B).

Fig. 4.

Efficacy of RET-TKI. A Response of RET-TKI in different application lines; B PFS of RET-TKI in different application lines; C Response of different therapeutic strategies in the first-line setting; D PFS of different therapeutic strategies in the first-line setting. ORR, objective response rate; RET-TKI, the tyrosine kinase inhibitor for rearranged during transfection gene; MKI, multikinase inhibitors; PD, progressive disease; CR, Complete response; SD, stable disease; PR, partial response

In particular, in the first-line setting, the clinical efficacy of RET-TKI was compared with that of other treatment strategies. The ORR of the RET-TKI group was 73.1% (19/26), which was higher than that of the other treatment strategies (P = 0.002), of which the ORR was 23.5% (4/17) in the “chemo-based” group, 50.0% (1/2) in the “chemo-free” group and 0.0% (0/2) in the “MKI” group (Fig. 4C). The mPFS was 22.7 months (95%CI, 12.9–32.4) for the “RET-TKI” group, 4.9 months (95%CI, 0–10.3) for the “chemo-based” group, 4.0 months (95%CI, NR-NR) for the “chemo-free” group and 6.6 months (95%CI, NR-NR) for the “MKI” group (Fig. 4D). RET-TKI had significantly prolonged PFS compared to other regimens (vs. chemo-based: HR = 5.1, P = 0.000; vs. chemo-free: HR = 10.9, P = 0.004; vs. MKI: HR = 5.2, P = 0.037).

Results of different fusion partners

The impact of fusion partners on efficacy of RET-TKI was evaluated. The ORRs for KIF5B and non-KIF5B were 58.6% and 76.9% respectively (P = 0.314), while the ORRs for CCDC6 and non-CCDC6 were 71.4% and 62.9% respectively (P = 1.000). There was a trend toward longer PFS in non-KIF5B patients than KIF5B patients (mPFS, 14.7 vs NR months, P = 0.081), and no difference was found between the PFS of CCDC6 patients and non-CCDC6 patients (mPFS, NR vs 16.1 months, P = 0.321) (Fig. 5A, B). Interestingly, the rate of baseline brain metastasis was higher in KIF5B patients (10/33, 30.3%) than in non-KIF5B patients (0/13, 0%) (P = 0.042). Among patients without baseline brain metastasis and received RET-TKI as first-line therapy, the rate of central nerves system (CNS) spread was 50% in both KIF5B (3/6) and non-KIF5B (2/4) patients.

Fig. 5.

Kaplan–Meier curves for PFS of RET-TKI of (A) KIF5B and non-KIF5B fusions; B CCDC6 and non-CCDC6 fusions; C different ECOG PS groups; D The resistance mode of first-line RET-TKI and the treatment after progression. PFS, progression-free survival; RET-TKI, the tyrosine kinase inhibitor for rearranged during transfection gene. CT, chemotherapy; RT, radiotherapy; AA, antiangiogenic therapy

Results of different ECOG PS scores

The efficacy of RET-TKI was evaluated in patients with different performance status. The ORR showed no statistical difference among the PS groups (P = 0.588), being 72.7%, 61.9% and 50.0% for patients with ECOG PS 0, 1 and ≥ 2, respectively. However, the PFS decreased as PS score increased, with the mPFS being 28.0 months (95%CI, 12.2–43.7), 16.1 months (95%CI, 11.8–20.3) and 7.5 months (95%CI, 5.7–9.2) for PS scoring 0, 1 and ≥ 2 respectively (P = 0.018) (Fig. 5C).

Intracranial efficacy of RET-TKI

The baseline brain metastasis rate was 19.6% (10/51) in the entire cohort. In patients with baseline brain metastasis (n = 10), the RET-TKI was used as first-line therapy in 5 patients, second-line therapy in 3 patients and third-line therapy or above in 2 patients. Systemic treatment prior to RET-TKI included four chemotherapies and one MKI. The RET-TKI used was pralsetinib. The intracranial response was observed in five patients (50%) and the intracranial DCR was 100%. The median duration of response was 16.4 months (95%CI, NR-NR) and the median intracranial PFS was 19.2 months (95%CI, 2.9–29.8). Among patients without baseline brain metastasis (n = 41), six (14.6%) progressed to the brain during the application of RET-TKI.

The patterns of progression on the first-line RET-TKI

The analysis of progression patterns was limited to patients who progressed on first-line RET-TKI (n = 13). Among these 13 patients, 3 had baseline brain metastasis, with one of which experiencing brain progression. Of the remaining 10 patients without baseline brain metastasis, 5 developed brain metastases during progression. Thus, progression in the brain accounted for 46.2% (6/13), and half of these patients continued RET-TKI, either alone or with additional treatments (including radiotherapy), after brain progression. Similarly, RET-TKI were also reserved for patients with liver progression, accounting for 15.4% (2/13). For patients with pleural (3/13, 23.1%), lung (1/13, 7.7%) or primary tumor (2/13, 15.4%) progression, the treatment strategies were usually switched to chemotherapy (Fig. 5D, Table 2). The ORR of second-line treatment was 20% and the mPFS was 2.7 months (95%CI, 1.9–3.5). Rebiopsies were scarcely conducted in case of progression due to lack of feasibility. In this case, liquid biopsies through blood tests were more often chosen than tissue biopsies, however, yielding limited information: only one patient had a low-abundance RET rearrangement detected in liquid biopsy.

Table 2.

The patterns of progression on the first-line RET-TKI

Patterns of progression	Number (%)
Intrapulmonary
Primary tumor	2 (15.4)
Lung metastasis	1 (7.7)
Extrapulmonary
Pleural metastasis	3 (23.1)
Brain metastasis	6 (46.2)
Liver metastasis	2 (15.4)

Safety of RET-TKI

Treatment-related adverse events (AEs) were analyzed in 35 patients, and any grade of AEs was reported in 29 (82.8%) of them. Treatment-related AEs of grade 3 or worse were observed in 8 (22.8%) patients (Table 3). The most common AEs of grade 3 or worse were decreased neutrophil count and anemia (11.4%). Treatment-related AEs led to 6 (17.1%) dose modification or treatment discontinuation events, five of which could be attributed to myelosuppression. No grade 5 AEs related to treatment were observed.

Table 3.

Adverse events

Treatment-related adverse events	Any grade (N = 29)	Grade ≥ 3 (N = 8)
Decreased neutrophil count	11 (31.4%)	4 (11.4%)
Anemia	8 (22.9%)	4 (11.4%)
Increased blood pressure	6 (17.1%)	0
Abnormal hepatic function	6 (17.1%)	0
Edema	5 (14.3%)	0
Fatigue	4 (11.4%)	0
Decreased platelet count	3 (8.6%)	3 (8.6%)
Pneumonitis	3 (8.6%)	1 (2.9%)
Constipation	3 (8.6%)	0
Dry mouth	2 (5.7%)	0
Fever	2 (5.7%)	0
Nausea	2 (5.7%)	0
Paresthesia	1 (2.9%)	0
Toothache	1 (2.9%)	0
Short of breath	1 (2.9%)	0
Decreased appetite	1 (2.9%)	0
Hyperuricaemia	1 (2.9%)	0

Discussion

This retrospective chart review compensates for the lack of clinical data in China by revealing the efficacy and safety of RET-TKI in patients with RET-rearranged advanced NSCLC in a real-world context.

Based on evidence from other driver genes, small-molecular targeted therapy should be better to be applied at the front line. However, in real-world practice, RET-TKI was commonly applied at later lines because of their cost and drug accessibility. Therefore, we evaluated the influence of different application lines on the clinical efficacy of RET-TKI and found that first-line administration had a numerically higher ORR and PFS than second-line treatment or above; however, the difference was not statistically significant. According to the LIBRETTO-001 trial, a phase I/II registrational trial evaluating the safety and efficacy of selpercatinib for RET-rearranged NSCLC, the mPFS of selpercatinib in patients with previous platinum chemotherapy was 24.9 months (95%CI, 19.3-NE), which was no shorter than that in treatment-naïve patients, which was 22.0 months (95%CI, 13.8-NE) [17]. Similarly, in phase I/II ARROW study evaluating the safety and efficacy of pralsetinib, the mPFS of treatment-naïve patients was 13.0 months (95%CI, 9.1-NR), whereas the mPFS of patients with prior platinum-based chemotherapy was 16.5 months (95%CI, 10.5–24.1). These data suggested that the RET-TKI might attain satisfactory effects even when used at later lines.

In view of the above results, it was of clinical value to find out how RET-TKI performs compared to other treatment strategies in the first-line setting. In the present study, compared with other treatment regimens, RET-TKI was associated with a greater disease response rate and a 5-fold longer mPFS, which is consistent with the findings of clinical trials [17–19], indicating that the RET-TKI should be the first-line choice in patients with RET rearrangement. In addition, chemotherapy-based strategies played an important role in real-world first-line practice. Therefore, in this study, we conducted an indirect comparison between RET-TKI with chemotherapy-based strategies at first-line setting and found that RET-TKI was superior to chemotherapy-based regimens in terms of both clinical response and PFS, consistent with the LIBRETTO-431 trial, which compared first-line selpercatinib with platinum-based chemotherapy with or without pembrolizumab [19, 20]. Other strategies including chemo-free and MKI had limited efficacy in our study as were reported in previous studies so they should not be considered as the first-line treatment unless exceptional circumstances exist.

Whether fusion partner could influence the efficacy of RET-TKI was unknown. The data in this study revealed a trend toward poorer PFS in patients with KIF5B-RET fusion than non-KIF5B fusion. Besides, higher rate of baseline brain metastasis was observed in KIF5B than in non-KIF5B group. Although the CNS spread rate in patients without baseline brain metastasis was similar in KIF5B and non-KIF5B subgroup, large sample data was needed to explore whether patients with KIF5B have more propensity for CNS spread and poor prognosis.

Patients with poor performance status had shorter PFS and tended to have a lower response rate to RET-TKI, which implied that RET-TKI should be used in the early phase before the deterioration of patient’s physical condition.

The treatment options for patients who were resistant to first-line RET-TKI were limited, as was their treatment efficacy. The combination of RET-TKI and other therapies was considered mostly for brain and liver metastases, and in other situations, chemotherapy was most commonly chosen. Few patients underwent genetic testing after developing TKI resistance in this study; thus, no molecular resistance mechanism was evaluated.

The intracranial response of patients with baseline brain metastasis in our study (50%) was numerically lower than that reported in clinical trials, with an intracranial ORR of 70% (7/10) for pralsetinib and 85% for selpercatinib [17, 18]. In addition, in patients without baseline brain metastasis, the rate of brain progression after RET-TKI treatment was 14.6% in the current study, whereas the rate was lower than 1% in the ARROW trial [18]. Intracranial progression accounted for nearly 50% of relapse after first-line RET-TKI. Given the limited sample size and different patient constitutions of these studies, the results should be interpreted carefully. Pralsetinib has been reported to be able to penetrate the blood–brain barrier (BBB) and has a high intracranial response rate in clinical trials, however, responses within real-world settings could be more complicated. Given the high prevalence rate (approximately 46%) of brain metastasis during the disease course of RET-rearranged NSCLC, the regular clinical and imaging surveillance of central nervous system disease should not be waived even with high-BBB-penetrating targeted therapies, and more data are needed to support the intracranial efficacy in a real-world context.

The most common AE that caused treatment adjustment was hematological toxicity. No new AEs were found in the safety analysis. A randomized phase III trial compared the efficacy and safety of first-line selpercatinib with platinum-based chemotherapy with or without pembrolizumab and the results verified the superiority of selpercatinib over the combination of chemotherapy and pembrolizumab in terms of both systemic (HR = 0.46, P < 0.001) and intracranial efficacy (HR = 0.28) [19, 20]. However, selpercatinib had a higher incidence of increase in liver enzyme, hypertension and prolonged QT interval than the control group, and patients in East Asia were reported to have a higher incidence of ≥ grade 3 adverse events than other patients [19], indicating that the application of selpercatinib in Chinese patients should be more cautious and under appropriate clinical support.

This study has several limitations. First, the sample size was relatively small and the conclusions should be validated in the future. Second, the survival data had some bias because of the retrospective nature of this study. However, this study was based on a real-world setting and thus reflected the real-world treatment conditions in current clinical practice. Third, different test methods including PCR and NGS were included in the study, which may lower the reliability of the results. The extrapolation of the findings may therefore be affected by these limitations.

Conclusion

The RET-TKI had favorable efficacy and safety for advanced NSCLC patients with RET rearrangement in a real-world setting and were superior to other treatment strategies, such as chemotherapy-based regimens in the first line. The use of RET-TKI in later lines may also have corresponding clinical benefits. Brain metastases should be carefully monitored even during RET-TKI treatment due to the considerable risk for intracranial progression.

Authors’ contributions

Conceptualization, LSY, LY, HYX and YW; Data curation, LYT, XSH, XZH, JLL and YNY; Formal analysis, LSY and LYT; Methodology, LSY, LYT, LY and HYX; Project administration, YW; Software, YNY; Supervision, YW; Validation, YNY; Writing – original draft, LSY and LY; Writing – review, editing and revising, LSY, LYT, XSH, XZH, JLL, HYX, and YW. All authors have read and agreed to the published version of the manuscript.

Funding

The study received no funding.

Data availability

The raw data supporting the conclusions of this article will be made available by the authors on request.

Declarations

Ethics approval and consent to participate

The Ethics Committee of National Cancer Center/ Cancer Hospital/ Chinese Academy of Medical Sciences and Peking Union Medical College approved the study and waived the informed consent for its retrospective noninvasive nature (Approval Number: 18–070/1648).

Consent for publication

All authors have agreed the publication of this study.

Competing interests

The authors declare no competing interests.