To the Editor: With the widespread application of immune checkpoint inhibitors (ICIs) in clinical practice, acquired resistance (AR) to programmed death-1 (PD-1)/programmed death ligand 1 (PD-L1) axis inhibitors in advanced non-small cell lung cancer (NSCLC) develops and limits the durability of immunotherapy. AR is a clinical condition in which a patient initially responds to ICIs but relapses and progresses after a period of time. The rates of AR have not been routinely reported in lung cancer but typically range from 12.9% to 64%.[1] However, studies on subsequent management strategies and the mechanism of AR to PD-1/PD-L1 axis inhibitors in lung cancer are limited. The main exploratory strategies are immunotherapy rechallenge and the combination of ICIs with other therapies, including local therapy, chemotherapy, antiangiogenetic treatment, and cytotoxic T lymphocyte antigen 4 (CTLA-4) inhibitors.[2] The purpose of this study was to characterize the clinical patterns of AR and compare different subsequent therapies and outcomes in NSCLC patients after AR to PD-1/PD-L1 inhibitors. Moreover, we explored the underlying mechanism of AR by analyzing the alterations in next-generation sequencing (NGS) data before and after AR.

We reviewed all medical records of NSCLC patients who received PD-1/PD-L1 inhibitor treatment between January 1, 2017 and December 31, 2021 at Xinqiao Hospital, the Second Affiliated Hospital of the Third Military Medical University. Patient’s physical conditions were followed up by phone and the last follow-up date was July 31, 2022. The median follow-up was 16.6 months. AR was defined as progression of disease after an initial response to ICIs.[1] Patients met the following criteria: (1) pathologically confirmed NSCLC; (2) patients benefited from prior PD-1/PD-L1 inhibitor treatment for at least 3 months (four cycles of ICIs) before AR; and (3) patients had at least one measurable lesion at baseline and completed tumor response evaluation after the initiation of PD-1/PD-L1 axis inhibitor treatment at least once. Patients who had progressive disease (PD) during the first course of ICI treatment were excluded to avoid primary resistance. Tumor responses to ICIs were assessed using the Response Evaluation Criteria in Solid Tumors (RECIST) v1.1. This retrospective study was approved by the institutional review board of Xinqiao Hospital, Third Military Medical University, China (No: 2023-YANDI187-01). The requirement for written informed consent was waived.

The median time to AR was 8.1 months (range from 3.3 months to 31.7 months). To evaluate the effectiveness of subsequent treatment strategies after AR, we categorized patients receiving different therapeutic methods as four groups: immunotherapy alone group (I-alone), immunotherapy plus antiangiogenetic therapy group (I + A), immunotherapy plus chemotherapy group (I + C), and standard of care group (SOC). Overall survival (OS) was calculated from the date of disease progression after PD-1/PD-L1 axis inhibitor therapy until death or last follow-up. Progression-free survival (PFS) was measured based on the date from AR to disease progression or death after receiving subsequent treatment. Categorical variables were presented as number (%); continuous variables were presented as median with range. PFS and OS curves were generated using the Kaplan–Meier method. Differences in PFS and OS among groups were assessed by using the log-rank test. Hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated using the Cox proportional hazards model. Subgroup analyses were performed for OS and PFS using a Cox proportional hazards model. SPSS 26.0 statistical software (SPSS Inc., Chicago, IL, USA) was used to perform all statistical analyses. To explore the mechanism of AR, Gene Ontology (GO) (https://geneontology.org/) and Kyoto Encyclopedia of Genes and Genomes (KEGG) (https://www.kegg.jp/kegg/pathway.html) enrichment analyses were conducted using R language (version 4.2.1 R Foundation, Vienna, Austria).

A total of 736 NSCLC patients were reviewed, and 94 (12.8%) patients with AR to PD-1/PD-L1 axis inhibitors were identified. Among them, 78 patients were enrolled in our study and 16 patients treated with radiotherapy which may confound the results were excluded. The characteristics of the 78 evaluable patients are listed in Supplementary Table 1, https://links.lww.com/CM9/B982. PD-1/PD-L1 axis inhibitors were used in 39 (50.0%) patients as first-line treatment and half of the patients as second-line or later treatment. The progression patterns of patients with AR to ICIs are shown in Supplementary Figure 1, https://links.lww.com/CM9/B982. Sixty-four (82.1%) patients developed disease progression at only one site, 13 (16.7%) patients had progression at two sites, and one (1.3%) patients had progression involving three sites. The most common site of progression was the lung (31 patients, 39.7%) followed by the bone (n = 10, 12.8%), brain (n = 8, 10.3%), and lymph nodes (LNs) (thoracic LNs and neck LNs, each n = 3 [3.8%], respectively). After AR to PD-1/PD-L1 axis inhibitors, 17 patients received immunotherapy alone (I-alone), 16 patients received immunotherapy plus antiangiogenetic therapy (I + A), 25 patients received immunotherapy combined with chemotherapy (I + C), and 20 patients received standard of care (SOC). As of July 31, 2022, 40 deaths and 62 patients with progressive diseases were reported, and the median overall (mOS) and median PFS (mPFS) time from AR to death or the last follow-up were 16.4 months and 7.0 months, respectively.

Several factors were analyzed to predict OS after AR. In multivariable regression analysis, the treatment strategy was an independent prognostic factor [Supplementary Table 2, https://links.lww.com/CM9/B982]. The OS and PFS of the four main sequential therapies after AR are shown in Supplementary Figure 2, https://links.lww.com/CM9/B982. Patients with I + A treatment and I-alone had better OS than patients with SOC treatment after AR to PD-1/PD-L1 axis inhibitors (OS: I + A vs. SOC, P = 0.002, I-alone vs. SOC, P <0.001). Patients with I + A treatment had significantly longer OS (P = 0.032) than patients with I + C treatment after AR to PD-1/PD-L1 axis inhibitors. The median PFS in the I+A treatment group was 11.70 months, longer than that of the I + C group (5.37 months), however, this disparity did not reach statistical significance (P = 0.065) [Supplementary Figure 3, https://links.lww.com/CM9/B982]. Furthermore, subgroup analyses showed that OS was better in I + A-treated patients with non-squamous histology, especially after second-line therapy, than in I + C-treated patients [Supplementary Figure 4, https://links.lww.com/CM9/B982].

To better understand the mechanism of AR, we analyzed next generation sequencing (NGS) data of baseline tumor biopsies or blood samples from patients. Notably, five patients underwent NGS before immunotherapy (NGS1) and after AR (NGS2), respectively. Interestingly, the results showed that mutated genes in patients after AR were enriched in the regulation of methylation and the pluripotency of stem cells [Supplementary Figure 5, https://links.lww.com/CM9/B982].

Our study showed longer OS in the I + A group than in the SOC group, which is consistent with the Lung-MAP S1800A trial that showed significant improvement of OS with ramucirumab and pembrolizumab vs. SOC in patients with advanced NSCLC previously treated with immunotherapy and chemotherapy.[3] Several previous studies reported that the combination of ICIs and antiangiogenic therapy normalizes vascular-immune crosstalk to enhance cancer immunity. We found that the I + A group had better OS than the I + C group. Gettinger et al[4] reported that most patients had only one site of progression, and the continuation of immunotherapy in addition to local therapy can yield an OS benefit in patients with oligoprogression. Similarly, we found that 64 (82.1%) patients developed only one site of progression, and the most common site of AR was the lung.

In lung cancer, the most remarkable mechanisms of AR are (1) aberrations in tumor neoantigen burden; (2) prevention of effector T-cell infiltration in the tumor microenvironment (TME) and T-cell exhaustion; (3) dysregulation of epigenetic modulation, signaling pathways, and transcriptional signatures in tumor; and (4) the dysbiosis of microbiome.[5] Karpf reported that epigenetic modification in cancer cells was related to immune-related genes, impacting antigen processing, presentation, and immune evasion. In our study, GO analysis was performed using mutated genes, and the results showed that methylation was highly enriched in NGS2.

There are two limitations in our study. One is the small number of patients, especially the patients with NGS data. Therefore, research with a large sample size is needed to screen out the optimal therapies for advanced NSCLC, not only to assess common combination treatment methods but also to explore other options, such as demethylating agents or other drugs. The other limitation is that this was a retrospective study from a single center. Collaborative efforts are needed to overcome the challenge of AR to PD-1/PD-L1 axis inhibitors and provide clinical benefits for patients.

In conclusion, we found that most patients developed progression at one organ (the lung) after AR to PD-1/PD-L1 axis inhibitors. For a large portion of advanced NSCLC patients, treatment with I + A after AR to ICIs resulted in significantly prolonged OS compared to that with I + C treatment after AR. These results may provide clinicians with effective therapeutic strategies to combat AR to ICIs in advanced NSCLC patients. Moreover, the enrichment of mutated genes in methylation and signaling pathways regulating the pluripotency of stem cells in this study may be a valuable clue to explore the mechanism of AR.

Funding

This work was supported by grants from the National Natural Science Foundation of China (No. 82173097 to Z.W.) and National Key Research&Development Program of China (Nos. 2022YFC2505000 and 2022YFC2505002 to H.L.).

Conflicts of interest

None.

References 1. Schoenfeld AJ, Antonia SJ, Awad MM, Felip E, Gainor J, Gettinger SN, et al. Clinical definition of acquired resistance to immunotherapy in patients with metastatic non-small-cell lung cancer. Ann Oncol 2021;32:1597–1607. doi: 10.1016/j.annonc.2021.08.2151. 2. Schoenfeld AJ, Hellmann MD. Acquired resistance to immune checkpoint inhibitors. Cancer Cell 2020;37:443–455. doi: 10.1016/j.ccell.2020.03.017. 3. Reckamp KL, Redman MW, Dragnev KH, Minichiello K, Villaruz LC, Faller B, et al. Phase II randomized study of ramucirumab and pembrolizumab versus standard of care in advanced non-small-cell lung cancer previously treated with immunotherapy-lung-MAP S1800A. J Clin Oncol 2022;40:2295–2306. doi: 10.1200/JCO.22.00912. 4. Gettinger SN, Wurtz A, Goldberg SB, Rimm D, Schalper K, Kaech S, et al. Clinical features and management of acquired resistance to PD-1 axis inhibitors in 26 patients with advanced non-small cell lung cancer. J Thorac Oncol 2018;13:831–839. doi: 10.1016/j.jtho.2018.03.008. 5. Sun Q, Wei X, Wang Z, Zhu Y, Zhao W, Dong Y. Primary and acquired resistance against immune check inhibitors in non-small cell lung cancer. Cancers (Basel) 2022;14:3294. doi: 10.3390/cancers14143294.