A retrospective analysis of immunotherapy in non-small cell lung cancer patients with targetable genetic mutations

Shiyu Zhang; Jingwen Liu; Jia Liu; Kaihua Lu

doi:10.1097/MD.0000000000044144

. 2025 Sep 12;104(37):e44144. doi: 10.1097/MD.0000000000044144

A retrospective analysis of immunotherapy in non-small cell lung cancer patients with targetable genetic mutations

Shiyu Zhang ^a, Jingwen Liu ^b, Jia Liu ^c, Kaihua Lu ^b,^*

PMCID: PMC12440478 PMID: 40958347

Abstract

Targeted therapy has greatly improved the survival time and quality of life of some patients with non-small cell lung cancer. The application of immunotherapy among populations without targetable oncogenes has also achieved great success. However, the decision is still inconclusive as to whether immunotherapy can be used in targeted populations. We screened studies related to immunotherapy for lung cancer from public data platforms and then rearranged the cases. We grouped these cases according to driver genes and analyzed the impact of different targetable genes on immunotherapy (including mono-immunotherapy and combined immunotherapy). In addition, we also identified the predictive efficacy of programmed death-ligand 1 and tumor mutation burden on immunotherapy in populations with different driver genes. We identified 926 cases of lung cancer doing immunotherapy, including 321 cases in the driver gene negative group, 289 cases in the non-sensitive mutations group, and 316 cases in the sensitive mutations group. Except for KRAS and BRAF-sensitive mutations, the other sensitive mutations seemed to impede the effectiveness of immunotherapy. While among the non-sensitive mutations, aside from the EGFR group, the others exhibited immune benefits. The predictive effects of programmed death-ligand 1 and tumor mutation burden on immunotherapy were no longer uniform in mono or combined immunotherapy and needed to be discussed separately. Our data suggest that certain driver gene mutations may influence the response to immunotherapy, though further validation in larger cohorts is needed to confirm these observations. Combined immunotherapy is also different from mono-immunotherapy in many aspects, so we need to consider it separately according to different situations in clinical treatment.

Keywords: immunotherapy, non-small cell lung cancer, targetable oncogenes

1. Introduction

Lung cancer is currently one of the most common cancers globally, with high incidence and mortality rates among malignant tumors.^[1] According to histological classification, lung cancer can be roughly divided into non-small cell lung cancer (NSCLC, approximately 85%) and small cell lung cancer (approximately 15%). Among NSCLC, lung adenocarcinoma has become the predominant type with a proportion of 78%.^[2] The discovery of lung cancer-related oncogenic driver genes has advanced the development of targeted therapies, greatly improving the survival time and quality of life for some NSCLC patients, particularly those with lung adenocarcinoma.^[3–6] The emergence of immune checkpoint inhibitors (ICIs) has made up for the treatment gap for patients without targetable oncogene alterations and significantly improved their therapeutic effect.^[7–9] Currently, molecular testing and programmed death-ligand 1 (PD-L1) testing have become the standard examinations for advanced NSCLC patients to assess suitability for targeted therapy or immunotherapy. Molecular profiling according to the National Comprehensive Cancer Network Clinical Practice Guidelines in oncology (NCCN guidelines) includes EGFR, KRAS, ALK, ROS1, BRAF, NTRK1/2/3, MET, RET, and ERBB2. Patients with negative molecular markers mentioned above are advised to use immunotherapy, using PD-L1 and tumor mutation burden (TMB) to predict the effectiveness of ICI.^[10]

During practical clinical treatment, we often encounter patients with driver oncogenes who are unable to continue targeted therapy due to poor response to targeted drugs, significant side effects, inability to afford targeted drugs, or acquired resistance. In such cases, they face the question of whether they can undergo immunotherapy. Although much research has been reported on whether patients with driver oncogenes could benefit from immunotherapy, there is still no consensus.^[11–14] It is also unknown whether the response to immunotherapy differs between patients with acquired resistance to targeted therapy and other targetable treatment populations. The monotherapy targeting programmed cell death protein 1 (PD-1) or PD-L1 has been widely used, while the combination therapy of anti-PD-(L)1 plus anti-cytotoxic T-lymphocyte-associated protein 4 (anti-CTLA-4) has further opened up new prospects for immunotherapy.^[15–17] Will patients with targetable oncogenes benefit differently from mono-immunotherapy or combination immunotherapy? We often use PD-L1 and TMB to predict the efficacy of ICIs but will they still work when it comes to patients with targetable oncogenes? Furthermore, we are concerned about a group of patients who carry gene alterations in sites similar to targetable driver genes but lack definitive evidence for targeted therapy, such as EGFR fusion mutations, KRAS G12V, KRAS G13D, etc. Will these patients have similar responses to immunotherapy as patients with corresponding targetable oncogenes? When deciding on a treatment regimen, we tend to consider them alongside patients who are negative for driver oncogenes. Is this appropriate? Based on all these questions, we hope to find some advanced NSCLC patients with different targetable oncogenes who have undergone immunotherapy and figure out if they can benefit from it, thus guiding future clinical treatment directions.

2. Methods

2.1. Data filtering and grouping

As shown in the data filtering flowchart (Fig. 1A), we searched for studies on lung cancer immunotherapy through the cBioPortal for Cancer Genomics website (https://www.cbioportal.org/) and selected cases for analysis. The selection criteria were as follows: advanced NSCLC cases; patients who received only immunotherapy at a certain stage of the disease. The treatment can be anti-PD-(L)1 or anti-PD-(L)1 combined with anti-CTLA-4, but without concurrent other anti-tumor treatments; cases with molecular testing information; cases with survival information: progression-free survival (PFS) or overall survival (OS). According to the 2023 NSCLC NCCN guideline (Version 4.2023), which lists 9 molecular biomarkers: EGFR, KRAS, ALK, ROS1, BRAF, NTRK1/2/3, MET, RET, and ERBB2, we divided the cases into 9 alteration groups and 1 negative group that does not include the alterations above.

Figure 1. — Data filtering and grouping. (A) Data screening flowchart. NCSLC = non-small cell lung cancer. (B) Patients grouping.

The OncoKB website (MSK’s Precision Oncology Knowledge Base, https://www.oncokb.org/) categorizes tumor molecular alterations into 4 treatment levels and 2 resistance levels: treatment level 1 (T1): Food and Drug Administration (FDA)-recognized biomarker predictive of response to an FDA-approved drug in this indication; level 2 (T2): standard care biomarker recommended by the NCCN or other professional guidelines predictive of response to an FDA-approved drug in this indication; level 3 (T3): compelling clinical evidence supports the biomarker as being predictive of response to a drug in this indication; level 4 (T4): compelling biological evidence supports the biomarker as being predictive of response to a drug. Resistance level 1 (R1): standard care biomarker predictive of resistance to an FDA-approved drug in this indication; level 2 (R2): compelling clinical evidence supports the biomarker as being predictive of resistance to a drug. We categorized molecular alterations in T1–2 as sensitive groups to existing targeted drugs, and T3–4 as non-sensitive groups. Alterations categorized as R1 were classified as the resistant group. Based on this, we further divided the 9 alteration groups above into sensitive and non-sensitive groups.

For cases carrying more than 1 molecular alteration, priority was given to whether they contained the sensitive one. If there was one or more sensitive alterations present, it was classified into the corresponding sensitive group. Cases without sensitive alterations were classified into the respective non-sensitive groups. Therefore, some cases may overlap between sensitive groups and between non-sensitive groups, but when we counted the number of cases and analyzed all non-sensitive cases as 1 group, we performed clinical information filtering and removed duplicate cases.

Although we selected cases from 6 different studies, considering that the patients involved may come from the same hospital, we filtered and removed samples with identical clinical information. For cases from the same study, if a patient had more than 1 sample, we kept only 1 sample (prioritizing samples with sensitive alterations if there were differences in the genetic information of these samples).

We also filtered patients with sensitive alterations from these 6 studies who did not undergo immunotherapy but received corresponding targeted therapy. We grouped these cases as a targeted therapy control for comparative analysis with immunotherapy group.

2.2. Data analysis

Survival analysis was performed using the Kaplan–Meier method through GraphPad Prism 9 (GraphPad Software, Version 9, Boston). Survival curves were compared using the Log-rank (Mantel–Cox) test. For the PD-L1 and TMB stratification analysis part, we first sorted the cases in each group based on PD-L1/TMB expression levels and then used the quartile method for stratification.

3. Results

3.1. Data filtering and grouping

We obtained a total of 6 studies (Table S1, Supplemental Digital Content, https://links.lww.com/MD/P796), from which we further identified 926 patients with advanced NSCLC who underwent immunotherapy alone (Fig. 1A and B). A total of 321 patients were grouped in the targetable oncogenes negative group and 289 patients were identified with the non-sensitive alterations. The sensitive groups included EGFR (n = 84), KRAS (n = 145), ALK (n = 3), ROS1 (n = 12), BRAF (n = 10), MET (n = 27), RET (n = 2), and ERBB2 (n = 33), and sensitive alterations for NTRK1/2/3 were not found. As to resistance groups, we identified 2 kinds of EGFR resistance alteration: T790M mutation (n = 14) and EGFR exon 20 insertion (n = 15), both indicating resistance to first-generation EGFR-TKIs (erlotinib, afatinib, gefitinib).

3.2. Immunotherapeutic efficacy among different alterations

Taking the driver oncogenes negative cases as a control, we analyzed whether the sensitive and non-sensitive groups benefit from immunotherapy. From the perspective of PFS analysis (Fig. 2A), the EGFR-sensitive group and ALK-sensitive group showed immunotherapy progression (Figure S1, Supplemental Digital Content, https://links.lww.com/MD/P795). Since the limited PFS information of ALK-sensitive group (only 1 case), we did not include it in Figure 2. On the other way, the BRAF and NTRK1/2/3 non-sensitive groups benefited from immunotherapy. Although not statistically significant, the other non-sensitive groups showed a tendency towards immunotherapy benefit except for the EGFR, which tended towards immunotherapy progression.

Figure 2. — Forest map of immunotherapy efficacy. PFS (A) and OS (B) of immunotherapy among different groups. EGFR-N = non-sensitive alteration of EGFR, the other non-sensitive groups were expressed identically. OS = overall survival, PFS = progression-free survival.

In terms of OS analysis (Fig. 2B), we found that 3 groups were contradictory to their PFS results (EGFR-sensitive group, ALK-sensitive group, and MET-sensitive group). The EGFR-sensitive group, which showed immunotherapy progression in terms of PFS, had OS results similar to the negative group and did not exhibit hyper-progression. Considering that targeted therapy is the current standard treatment for patients with corresponding mutations, we further compared the OS between the EGFR-sensitive group in immunotherapy and targeted therapy (Figure S2a, Supplemental Digital Content, https://links.lww.com/MD/P795) and found that the targeted therapy group had significantly longer OS than the immunotherapy group. Thus, we speculate that patients in the immunotherapy group might have received targeted therapy at other times during the disease course, thereby prolonging their OS.

Similar to the EGFR-sensitive group, the ALK-sensitive group showed superior OS in targeted therapy compared to immunotherapy (Figure S2b, Supplemental Digital Content, https://links.lww.com/MD/P795), possibly due to the use of targeted drugs prolonging the OS of the immunotherapy group, which did not show a tendency towards immunotherapy progression.

The MET-sensitive group showed a tendency towards immunotherapy progression in terms of PFS but had better OS than the negative group. We compared the OS between MET-sensitive patients treated with targeted therapy and immunotherapy, and although the survival curve of the targeted therapy group was above the immunotherapy group, the results were not statistically significant (Figure S2c, Supplemental Digital Content, https://links.lww.com/MD/P795). Therefore, it is currently uncertain whether the extended OS in the immunotherapy group is due to the use of targeted drugs or if they indeed benefited from immunotherapy.

Additionally, the BRAF-sensitive group (i.e., patients with BRAF V600E mutation) showed immunotherapy progression in terms of OS, while the ALK, NTRK1/2/3, MET and ERBB2 non-sensitive groups showed immunotherapy benefit.

3.3. Efficacy difference between anti-PD-(L)1 and anti-PD-(L)1 + anti-CTLA-4

In this study, about 85% of the patients (n = 788) received anti-PD-(L)1 mono-immunotherapy, and 15% (n = 138) received anti-PD-(L)1 + anti-CTLA-4 combination immunotherapy (Table 1). Therefore, we further analyzed their efficacy difference.

Table 1.

Clinical characteristic of the patients.

Characteristic	No. (%)
No. of patients	926
Median age (range)	67 (22–93)
Gender
Male	450 (49)
Female	476 (51)
Histology
Adenocarcinoma	799 (86)
Squamous	83 (9)
Other	44 (5)
Treatment
PD-(L)1, monotherapy	788 (85)
PD-(L)1 + CTLA-4, combination therapy	138 (15)
PD-L1 expression
0	194 (21)
1–49%	112 (12)
≥50%	112 (12)
NA	508 (55)
TMB
<10 mutations/Mb	630 (68)
≥10 mutations/Mb	296 (32)

Open in a new tab

CTLA-4 = cytotoxic T-lymphocyte-associated protein 4, PD-L1 = programmed death-ligand 1, TMB = tumor mutation burden.

For the negative group, combination therapy showed a significant improvement in PFS compared to monotherapy (Fig. 3A). Although the data on OS for combination therapy were limited (8 cases), they also demonstrated better results than monotherapy (Figure S3a, Supplemental Digital Content, https://links.lww.com/MD/P795).

Figure 3. — PFS difference between anti-PD-(L)1 and anti-PD-(L)1 + anti-CTLA-4. (A) PFS of the negative group; (B) PFS of EGFR-sensitive and non-sensitive groups; (C) PFS of KRAS-sensitive and non-sensitive groups; (D) PFS of ALK-sensitive and non-sensitive groups; (E) PFS of ROS1 non-sensitive group; (F) PFS of NTRK1/2/3 non-sensitive group; (G) PFS of MET-sensitive and non-sensitive groups; (H) PFS of RET non-sensitive group; (I) PFS of the non-sensitive population. PD-1 = programmed cell death protein 1, PD-L1 = programmed death-ligand 1, PFS = progression-free survival.

For patients with EGFR mutations, the sensitive group in both monotherapy and combination therapy and the non-sensitive group in monotherapy all showed immune progression. Only the non-sensitive group might benefit from combination therapy (Fig. 3b). However, due to the limited number of patients in the non-sensitive group receiving combination therapy, the result had no statistical significance. Furthermore, there was no OS information for the EGFR non-sensitive group with combination therapy (Figure S3b, Supplemental Digital Content, https://links.lww.com/MD/P795), which hindered further assessment.

In both the KRAS-sensitive and non-sensitive groups, there was no significant difference in PFS between monotherapy and combination therapy. However, from the graph, the PFS curve for the KRAS non-sensitive group in combination therapy was positioned at the top (Fig. 3C), suggesting that further comparisons and analysis with a larger sample size may be warranted in clinical practice. There was no difference in OS between monotherapy and combination therapy for both the KRAS-sensitive and non-sensitive groups, and although the survival curve for combination therapy was above that of monotherapy, it was not statistically significant due to the limited number of cases (Figure S3c, Supplemental Digital Content, https://links.lww.com/MD/P795).

In the ALK-sensitive group, there was only 1 case of monotherapy, which showed PFS immune progression. There was no statistical difference in PFS between monotherapy and combination therapy in the ALK non-sensitive group (Fig. 3D). OS information was lacking for combination therapy, preventing further analysis.

For the ROS1 group, there was a lack of information on combination therapy in the sensitive group. No significant difference was found in PFS between monotherapy in both the sensitive and non-sensitive groups (Fig. 3E). OS information was lacking for combination therapy, preventing further analysis.

In the NTRK1/2/3 non-sensitive group, there were no significant differences in PFS or OS between monotherapy and combination therapy (Fig. 3F, Figure S3d, Supplemental Digital Content, https://links.lww.com/MD/P795). There were no significant differences in PFS or OS between monotherapy and combination therapy for both the MET-sensitive and non-sensitive groups (Fig. 3G, Figure S3e, Supplemental Digital Content, https://links.lww.com/MD/P795).

There was no information available for the RET-sensitive group, and in the non-sensitive group, the combination therapy curve was above the monotherapy in terms of PFS analysis, but it may not have had statistical significance due to the limited number of cases (Fig. 3H). The RET lacked OS information and failed to be analyzed.

What’s more, we analyzed the non-sensitive alterations as a whole, and the results were similar to those in the negative group. The non-sensitive cases had better PFS and OS when using combination therapy than the monotherapy (Fig. 3I, Figure S3f, Supplemental Digital Content, https://links.lww.com/MD/P795).

3.4. Immunotherapeutic efficacy among resistant cases

Some driver oncogenes indicate resistance to certain targeted therapies. We analyzed whether patients with these alterations could benefit from immunotherapy. As shown in Figure 4, patients with EGFR T790M and EGFR exon 20 insertion did not show significant differences in PFS and OS when receiving immunotherapy compared to other EGFR-sensitive mutation patients, indicating that they did not benefit from immunotherapy.

Figure 4. — Immunotherapeutic efficacy among resistant cases. (A) PFS of the EGFR T790M, EGFR exon 20 insertion and the other EGFR alteration groups; (B) OS of the EGFR T790M, EGFR exon 20 insertion and the other EGFR alteration groups. OS = overall survival, PFS = progression-free survival.

3.5. Immunotherapeutic efficacy stratified by PD-L1

Based on the analysis of monotherapy and combination therapy above, we further conducted an analysis of the relationship between PFS and PD-L1 expression level. Considering the difference in efficacy between monotherapy and combination therapy in the negative group, we analyzed them separately according to the PD-L1 expression.

As shown in Figure 5A, for monotherapy of the negative group, higher PD-L1 expression was associated with better monotherapy efficacy, while for patients receiving combination therapy, PD-L1 expression level did not seem to predict efficacy (Figure S4a, Supplemental Digital Content, https://links.lww.com/MD/P795).

Since the limited number of cases, we failed to analyze the EGFR non-sensitive samples with combination therapy. Considering that the remaining cases with EGFR mutations could not benefit from immunotherapy, we did not perform further analysis.

The KRAS-sensitive and non-sensitive groups were analyzed together for they showed no significant difference in efficacy between monotherapy and combination therapy. Figure 5B indicated that PD-L1 could predict immunotherapy efficacy in KRAS patients, where a higher expression level was associated with better PFS.

Information for both the ALK-sensitive and non-sensitive groups was limited and could not be analyzed. The PD-L1 stratification in the ROS1 non-sensitive group showed no statistical significance (Figure S4b, Supplemental Digital Content, https://links.lww.com/MD/P795), while the sensitive group had limited PD-L1 information and could not be analyzed.

The BRAF group had no cases receiving combination therapy, thus the analysis was only performed for monotherapy. The PD-L1 stratification in the BRAF non-sensitive group showed no statistical significance (Figure S4c, Supplemental Digital Content, https://links.lww.com/MD/P795). The BRAF-sensitive group (BRAF V600E) had few cases and could not be analyzed.

Since the NTRK1/2/3 non-sensitive group showed no significant difference in efficacy between monotherapy and combination therapy, they were analyzed together. Figure 5C indicated that PD-L1 had no predictive ability for immunotherapy efficacy in such patients.

As to MET, sensitive and non-sensitive groups were also analyzed together due to the small efficacy difference between monotherapy and combination therapy. Figure 5D suggested that PD-L1 could effectively differentiate immunotherapy efficacy in this group of patients, where higher expression level was associated with better PFS.

There were few cases in the RET-sensitive group, while more than half of the samples in the non-sensitive group had PD-L1 expression as negative. Therefore, we grouped the patients based on PD-L1 negative or positive (greater than or equal to 1), and the results suggested that there was no significant relationship between immunotherapy efficacy and PD-L1 (Fig. 5E).

The PFS curves of the ERBB2-sensitive and the non-sensitive group were separated, so we first tried to analyze them separately (Figure S4d, S4e, Supplemental Digital Content, https://links.lww.com/MD/P795), and then we found that the trends of these 2 groups were similar, so we finally graphed them together. As shown in Figure 5F, the PD-L1 positive group with ERBB2 mutation had significantly better immunotherapy efficacy than the negative one.

3.6. Immunotherapeutic efficacy stratified by TMB

TMB stratified analysis did the same as the PD-L1 section above when deciding whether it was required to distinguish monotherapy or combination therapy. TMB could predict the efficacy of monotherapy in the negative population (Fig. 6A) but failed in patients undergoing combination immunotherapy (Figure S5a, Supplemental Digital Content, https://links.lww.com/MD/P795). For the KRAS population, the higher the TMB level, the better the immunotherapy efficacy (Fig. 6B).

Figure 6. — Immunotherapeutic efficacy stratified by TMB. (A) PFS of the negative group with anti-PD-(L)1 treatment stratified by TMB; (B) PFS of KRAS group stratified by TMB; (C) PFS of ALK non-sensitive group stratified by TMB; (D) PFS of ROS1 non-sensitive group stratified by TMB; (E) PFS of BRAF non-sensitive group stratified by TMB; (F) PFS of NTRK1/2/3 non-sensitive group stratified by TMB; (G) PFS of MET group stratified by TMB; (H) PFS of RET group stratified by TMB; (I) PFS of ERBB2 group stratified by TMB. PD-1 = programmed cell death protein 1, PD-L1 = programmed death-ligand 1, PFS = progression-free survival, TMB = tumor mutation burden.

There were little cases to analyze in the ALK-sensitive group. We analyzed single and combination immunotherapy in the ALK non-sensitive group together considering the little difference between them. Figure 6C suggested that TMB expression levels can predict the immune efficacy of ALK non-sensitive alteration patients, with higher TMB levels correlating with better efficacy.

TMB stratification in the ROS1-sensitive group showed no statistical significance (Figure S5b, Supplemental Digital Content, https://links.lww.com/MD/P795), while in the non-sensitive group, patients with higher TMB had significantly better PFS than the lower part (Fig. 6D).

Although TMB stratification in the BRAF non-sensitive group showed no statistical significance, from the figure, it could be seen that the efficacy of patients with TMB expression in the top 75% is better than the bottom 25% (Fig. 6E).

The immune efficacy of the NTRK1/2/3 non-sensitive group correlated with TMB levels. PFS was better in the top 75% TMB levels than in the bottom 25%, but there was no significant layer-by-layer variation between the top 75%, top 50%, and top 25% groups (Fig. 6F).

TMB could also predict the immune efficacy of the MET alteration population, as shown in Figure 6G, where the top 75% TMB group was better than the bottom 25%. The immunotherapy efficacy of the RET non-sensitive group was related to TMB, similar to the MET group, where the PFS of the bottom 25% was significantly worse than the top 75% (Fig. 6H).

There was no obvious stratification trend between the ERBB2-sensitive group and the non-sensitive group (Figure S5c and S5d, Supplemental Digital Content, https://links.lww.com/MD/P795), so similar to the PD-L1 analysis above, we put the 2 groups together. As shown in Figure 6I, TMB failed to predict the immunotherapy efficacy in this population.

4. Discussion

When patients with targetable oncogenes cannot use targeted therapy, whether they can use immunotherapy and how to use it becomes an unavoidable issue. Some previous studies have discussed the problem of immunotherapy for such patients,^[14,18–20] but many of these studies only involve 1 or 2 driver oncogenes and are almost limited to single immunotherapy. We took all 9 targetable oncogenes as well as combination immunotherapy into consideration in this research. Given the excellent efficacy of ICIs, immunotherapy is preferred as long as there are no contraindications for advanced NSCLC patients who do not carry targetable oncogenes.^[7] Therefore, unlike some studies that choose patients receiving chemotherapy alone as the control group or compare different mutation groups with each other,^[14,21] we chose the driver oncogene-negative population who received immunotherapy as the control. When the treatment effect is similar to or better than the negative group, it indicates that this population is a beneficiary of immunotherapy and recommends the use of immunotherapy drugs in clinical treatment.

Nowadays, most studies show that patients with EGFR mutations do not benefit from ICIs.^[21,22] As a result, many immunotherapy-related clinical studies have excluded patients carrying EGFR mutations.^[23,24] While some studies have driven the opposite conclusions that immunotherapy combined with other anti-tumor treatments can improve the survival time of the EGFR-positive population.^[25,26] However, these studies almost exclusively focus on patients with EGFR-sensitive mutations, and most do not differentiate between anti-PD-(L)1 or anti-PD-(L)1 + anti-CTLA-4. For patients with EGFR-sensitive mutations, our analysis found that they do not benefit from either anti-PD-(L)1 or anti-PD-(L)1 + anti-CTLA-4 treatment. As for patients with EGFR non-sensitive mutations, they may not use targeted drugs because they were not beneficiaries of targeted therapy according to the NSCLC guidelines. Therefore, whether they can benefit from immunotherapy is a question worth exploring. Our analysis suggested that patients with EGFR non-sensitive mutations also do not benefit from anti-PD-(L)1, but anti-PD-(L)1 + anti-CTLA-4 may still be an option for them in the future, which requires further research and exploration for validation.

Research has shown that the population with KRAS mutations tends to benefit more from immunotherapy due to a higher proportion of patients with high PD-L1 expression.^[14,20] It is important to note that this result is based on the comparison with other types of driver mutations. When we used the driver oncogene-negative population as the control, the results showed that the PFS curve of the KRAS population with immunotherapy almost overlapped with the negative group, suggesting that KRAS mutations seem to have little impact on the efficacy of immunotherapy. On the other hand, similar to published studies, our analysis also indicates that the mutation subtypes of KRAS have no relation to immunotherapy efficacy.

When we conducted the efficacy difference analysis of anti-PD-(L)1 and anti-PD-(L)1 + anti-CTLA-4, groups such as the non-sensitive groups of EGFR, KRAS, and RET did not have significant statistical significance due to the small number of cases. However, their survival curves suggested that these populations may still potentially benefit from the immunotherapy and this requires further exploration and validation in clinical practice.

In this study, we hoped to expand the number of cases as much as possible to observe the efficacy. The various non-sensitive groups in Figure 2A showed a consistent trend of immunotherapy benefit, even the exceptional EGFR non-sensitive group, which also showed a similar trend of benefit from combination immunotherapy when compared to monotherapy. Therefore, we combined the various non-sensitive groups, screened and removed duplicate information, and analyzed the efficacy of single and combination immunotherapy again. The results are similar to the control group. The population with non-sensitive alterations may potentially find hope for immunotherapy from anti-PD-(L)1 + anti-CTLA-4.

Some studies have suggested that patients with different oncogenes have different responses to immunotherapy due to their different levels of PD-L1 and TMB expression. Groups with poor immunotherapy efficacy, such as classical EGFR and EGFR exon 20 alteration, have lower PD-L1 and TMB expression, which consequently leads to poor therapeutic efficacy.^[12] On the other hand, groups with relatively better immunotherapy efficacy, such as non-sensitive BRAF, have more patients with high PD-L1 and TMB expression. Our study directly analyzed the correlation between PFS and the expression of PD-L1 and TMB of various oncogenes. This study suggested that PD-L1 and TMB are positively correlated with the efficacy of immunotherapy to a certain extent, such as in the population of the negative group, KRAS group, and MET-positive group. However, it is not all types of alterations can be predicted for their efficacy through PD-L1 and TMB. For some oncogene populations, there are differences in the actual predictive effects of PD-L1 and TMB. For example, PD-L1 had a significant ability to differentiate efficacy among patients with ERBB2 alteration, but TMB did not. For the RET group, TMB could predict the efficacy but PD-L1 did not show a clear correlation. For patients of non-sensitive BRAF and NTRK1/2/3 groups, neither PD-L1 nor TMB work well. What’s more, our study suggested that even when these 2 biomarkers predict the efficacy of single immunotherapy, they tended to fail in combination immunotherapy. One study suggests that molecular markers that can predict the efficacy of anti-PD-1 have no predictive ability for anti-CTLA-4,^[27] and this might be one of the explanations. Another research which includes cases of combination immunotherapy proposes that F2RL1 and RBFOX2 could serve as predictive targets for immunotherapy,^[28] but unfortunately, this study does not separate cases receiving combination or single immunotherapy. So, the predictive effects of the 2 targets are still unknown in anti-PD-(L)1 + anti-CTLA-4 treatment. With the clinical application of anti-PD-(L)1 + anti-CTLA-4, more research is needed to find other indicators to better predict the efficacy of combination immunotherapy.

Furthermore, there are still many shortcomings in this study. The study only included the cases of individuals who underwent immunotherapy alone during a certain period. However, other treatments during the disease are also important factors affecting OS. Therefore, we focused on PFS, using OS data as supplementary references. This means that this manuscript was unable to answer whether there is a synergistic or antagonistic effect between immunotherapy and targeted therapy. On the other hand, we only used PFS and OS as the evaluation criterion for treatment efficacy and did not comprehensively evaluate various aspects such as treatment efficacy and side effects. For the application of immunotherapy, future studies with larger, more diverse populations are still needed to achieve a more comprehensive understanding.

Acknowledgments

We are grateful for the collation and sharing of clinical information on the cBioPortal for Cancer Genomics website. We would like to thank Professor Yi Yang from the School of Life Science and Technology, Southeast University, for his statistical guidance.

Author contributions

Conceptualization: Shiyu Zhang, Kaihua Lu.

Data curation: Shiyu Zhang, Jingwen Liu.

Formal analysis: Shiyu Zhang.

Funding acquisition: Kaihua Lu.

Methodology: Shiyu Zhang.

Project administration: Shiyu Zhang, Kaihua Lu.

Visualization: Shiyu Zhang.

Validation: Jingwen Liu, Jia Liu.

Writing – review & editing: Jingwen Liu, Jia Liu, Kaihua Lu.

Writing – original draft: Shiyu Zhang.

Supplementary Material

medi-104-e44144-s001.docx^{(14.3KB, docx)}

medi-104-e44144-s002.docx^{(843.8KB, docx)}

Abbreviations:

CTLA-4: cytotoxic T-lymphocyte-associated protein 4
FDA: Food and Drug Administration
ICIs: immune checkpoint inhibitors
NCCN guidelines: the National Comprehensive Cancer Network Clinical Practice Guidelines in oncology
NSCLC: non-small cell lung cancer
OS: overall survival
PD-1: programmed cell death protein 1
PD-L1: programmed death-ligand 1
PFS: progression-free survival
TMB: tumor mutation burden

This research was funded by the National Natural Science Foundation of China, grant number 82172708.

This study did not require ethical approval since all the data are from public datasets.

The authors have no conflicts of interest to disclose.

The datasets generated during and/or analyzed during the current study are publicly available.The data used in this study were all sourced from the cBioPortal DataHub. The cBioPortal DataHub provides access to all publicly available datasets in a standardized format, supporting reproducible research and enabling new discoveries through data reuse.

Supplemental Digital Content is available for this article.

How to cite this article: Zhang S, Liu J, Liu J, Lu K. A retrospective analysis of immunotherapy in non-small cell lung cancer patients with targetable genetic mutations. Medicine 2025;104:37(e44144).

References

[1].Siegel RL, Miller KD, Wagle NS, Jemal A. Cancer statistics, 2023. CA Cancer J Clin. 2023;73:17–48. [DOI] [PubMed] [Google Scholar]
[2].Thai AA, Solomon BJ, Sequist LV, Gainor JF, Heist RS. Lung cancer. Lancet (London). 2021;398:535–54. [DOI] [PubMed] [Google Scholar]
[3].Planchard D, Jänne PA, Cheng Y, et al. Osimertinib with or without chemotherapy in EGFR-mutated advanced NSCLC. N Engl J Med. 2023;389:1935–48. [DOI] [PubMed] [Google Scholar]
[4].Shaw AT, Solomon BJ, Chiari R, et al. Lorlatinib in advanced ROS1-positive non-small-cell lung cancer: a multicentre, open-label, single-arm, phase 1-2 trial. Lancet Oncol. 2019;20:1691–701. [DOI] [PubMed] [Google Scholar]
[5].Camidge DR, Barlesi F, Goldman JW, et al. Phase Ib study of telisotuzumab vedotin in combination with erlotinib in patients with c-met protein-expressing non-small-cell lung cancer. J Clin Oncol. 2023;41:1105–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
[6].Tan AC, Tan DSW. Targeted therapies for lung cancer patients with oncogenic driver molecular alterations. J Clin Oncol. 2022;40:611–25. [DOI] [PubMed] [Google Scholar]
[7].Mok TSK, Wu YL, Kudaba I, et al. Pembrolizumab versus chemotherapy for previously untreated, PD-L1-expressing, locally advanced or metastatic non-small-cell lung cancer (KEYNOTE-042): a randomized, open-label, controlled, phase 3 trial. Lancet. 2019;393:1819–30. [DOI] [PubMed] [Google Scholar]
[8].Wu YL, Lu S, Cheng Y, et al. Nivolumab versus docetaxel in a predominantly Chinese patient population with previously treated advanced NSCLC: checkmate 078 randomized phase iii clinical trial. J Thorac Oncol. 2019;14:867–75. [DOI] [PubMed] [Google Scholar]
[9].Borghaei H, Paz-Ares L, Horn L, et al. Nivolumab versus docetaxel in advanced non-squamous non-small-cell lung cancer. N Engl J Med. 2015;373:1627–39. [DOI] [PMC free article] [PubMed] [Google Scholar]
[10].Ettinger DS, Wood DE, Aisner DL, et al. Non–small cell lung cancer, version 32022, NCCN clinical practice guidelines in oncology. J Natl Compr Canc Netw. 2022;20:497–530. [DOI] [PubMed] [Google Scholar]
[11].Dantoing E, Piton N, Salaün M, Thiberville L, Guisier F. Anti-PD1/PD-L1 immunotherapy for non-small cell lung cancer with actionable oncogenic driver mutations. Int J Mol Sci . 2021;22:6288. [DOI] [PMC free article] [PubMed] [Google Scholar]
[12].Negrao MV, Skoulidis F, Montesion M, et al. Oncogene specific differences in tumor mutationalburden,PD-L1 expression, and outcomes from immunotherapy in non- small cell lung cancer. J ImmunoTher Cancer. 2021;9:e002891. [DOI] [PMC free article] [PubMed] [Google Scholar]
[13].Qian C, Zhang Y, Cheng W, Zhang Q, Li M, Fang S. Case report: rechallenge with EGFR–TKIs after immunotherapy in EGFR–mutated non–small cell lung cancer with leptomeningeal metastasis. Front Oncol. 2022;12:957661. [DOI] [PMC free article] [PubMed] [Google Scholar]
[14].Mazieres J, Drilon A, Lusque A, et al. Immune checkpoint inhibitors for patients with advanced lung cancer and oncogenic driver alterations: results from the IMMUNOTARGET registry. Ann Oncol. 2019;30:1321–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
[15].Yap TA, Parkes EE, Peng W, Moyers JT, Curran MA, Tawbi HA. Development of immunotherapy combination strategies in cancer. Cancer Discov. 2021;11:1368–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
[16].Paz-Ares LG, Ramalingam SS, Ciuleanu TE, et al. First-line nivolumab plus ipilimumab in advanced NSCLC: 4-year outcomes from the randomized, open-label, phase 3 checkmate 227 part 1 trial. J Thorac Oncol. 2022;17:289–308. [DOI] [PubMed] [Google Scholar]
[17].Schoenfeld JD, Giobbie-Hurder A, Ranasinghe S, et al. Durvalumab plus tremelimumab alone or in combination with low-dose or hypofractionated radiotherapy in metastatic non-small-cell lung cancer refractory to previous PD(L)-1 therapy: an open-label, multicentre, randomised, phase 2 trial. Lancet Oncol. 2022;23:279–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
[18].Lisberg A, Cummings A, Goldman JW, et al. A phase II study of pembrolizumab in EGFR-mutant, PD-L1+, tyrosine kinase inhibitor Naïve patients with advanced NSCLC. J Thorac Oncol. 2018;13:1138–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
[19].Garassino MC, Cho BC, Kim JH, et al. Durvalumab as third-line or later treatment for advanced non-small-cell lung cancer (ATLANTIC): an open-label, single-arm, phase 2 study. Lancet Oncol. 2018;19:521–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
[20].Chen N, Fang W, Lin Z, et al. KRAS mutation-induced upregulation of PD-L1 mediates immune escape in human lung adenocarcinoma. Cancer Immunol Immunother. 2017;66:1175–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
[21].Lee CK, Man J, Lord S, et al. Checkpoint inhibitors in metastatic EGFR-mutated non-small cell lung cancer-a meta-analysis. J Thorac Oncol. 2017;12:403–7. [DOI] [PubMed] [Google Scholar]
[22].Qiao M, Jiang T, Liu X, et al. Immune checkpoint inhibitors in EGFR-mutated NSCLC: dusk or dawn? J Thorac Oncol. 2021;16:1267–88. [DOI] [PubMed] [Google Scholar]
[23].Reck M, Rodríguez-Abreu D, Robinson AG, et al. Five-year outcomes with pembrolizumab versus chemotherapy for metastatic non-small-cell lung cancer with PD-L1 tumor proportion score ≥ 50. J Clin Oncol. 2021;39:2339–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
[24].West H, McCleod M, Hussein M, et al. Atezolizumab in combination with carboplatin plus nab-paclitaxel chemotherapy compared with chemotherapy alone as first-line treatment for metastatic non-squamous non-small-cell lung cancer (IMpower130): a multicenter, randomized, open-label, phase 3 trial. Lancet Oncol. 2019;20:924–37. [DOI] [PubMed] [Google Scholar]
[25].Socinski MA, Jotte RM, Cappuzzo F, et al. Atezolizumab for first-line treatment of metastatic nonsquamous NSCLC. N Engl J Med. 2018;378:2288–301. [DOI] [PubMed] [Google Scholar]
[26].Nogami N, Barlesi F, Socinski MA, et al. IMpower150 final exploratory analyses for atezolizumab plus bevacizumab and chemotherapy in key NSCLC patient subgroups with EGFR mutations or metastases in the liver or brain. J Thorac Oncol. 2022;17:309–23. [DOI] [PubMed] [Google Scholar]
[27].Roh W, Chen PL, Reuben A, et al. Integrated molecular analysis of tumor biopsies on sequential CTLA-4 and PD-1 blockade reveals markers of response and resistance. Sci Transl Med. 2017;9:eaah3560. [DOI] [PMC free article] [PubMed] [Google Scholar]
[28].Bareche Y, Kelly D, Abbas-Aghababazadeh F, et al. Leveraging big data of immune checkpoint blockade response identifies novel potential targets. Ann Oncol. 2022;33:1304–17. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

medi-104-e44144-s001.docx^{(14.3KB, docx)}

medi-104-e44144-s002.docx^{(843.8KB, docx)}

[R1] [1].Siegel RL, Miller KD, Wagle NS, Jemal A. Cancer statistics, 2023. CA Cancer J Clin. 2023;73:17–48. [DOI] [PubMed] [Google Scholar]

[R2] [2].Thai AA, Solomon BJ, Sequist LV, Gainor JF, Heist RS. Lung cancer. Lancet (London). 2021;398:535–54. [DOI] [PubMed] [Google Scholar]

[R3] [3].Planchard D, Jänne PA, Cheng Y, et al. Osimertinib with or without chemotherapy in EGFR-mutated advanced NSCLC. N Engl J Med. 2023;389:1935–48. [DOI] [PubMed] [Google Scholar]

[R4] [4].Shaw AT, Solomon BJ, Chiari R, et al. Lorlatinib in advanced ROS1-positive non-small-cell lung cancer: a multicentre, open-label, single-arm, phase 1-2 trial. Lancet Oncol. 2019;20:1691–701. [DOI] [PubMed] [Google Scholar]

[R5] [5].Camidge DR, Barlesi F, Goldman JW, et al. Phase Ib study of telisotuzumab vedotin in combination with erlotinib in patients with c-met protein-expressing non-small-cell lung cancer. J Clin Oncol. 2023;41:1105–15. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R6] [6].Tan AC, Tan DSW. Targeted therapies for lung cancer patients with oncogenic driver molecular alterations. J Clin Oncol. 2022;40:611–25. [DOI] [PubMed] [Google Scholar]

[R7] [7].Mok TSK, Wu YL, Kudaba I, et al. Pembrolizumab versus chemotherapy for previously untreated, PD-L1-expressing, locally advanced or metastatic non-small-cell lung cancer (KEYNOTE-042): a randomized, open-label, controlled, phase 3 trial. Lancet. 2019;393:1819–30. [DOI] [PubMed] [Google Scholar]

[R8] [8].Wu YL, Lu S, Cheng Y, et al. Nivolumab versus docetaxel in a predominantly Chinese patient population with previously treated advanced NSCLC: checkmate 078 randomized phase iii clinical trial. J Thorac Oncol. 2019;14:867–75. [DOI] [PubMed] [Google Scholar]

[R9] [9].Borghaei H, Paz-Ares L, Horn L, et al. Nivolumab versus docetaxel in advanced non-squamous non-small-cell lung cancer. N Engl J Med. 2015;373:1627–39. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R10] [10].Ettinger DS, Wood DE, Aisner DL, et al. Non–small cell lung cancer, version 32022, NCCN clinical practice guidelines in oncology. J Natl Compr Canc Netw. 2022;20:497–530. [DOI] [PubMed] [Google Scholar]

[R11] [11].Dantoing E, Piton N, Salaün M, Thiberville L, Guisier F. Anti-PD1/PD-L1 immunotherapy for non-small cell lung cancer with actionable oncogenic driver mutations. Int J Mol Sci . 2021;22:6288. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R12] [12].Negrao MV, Skoulidis F, Montesion M, et al. Oncogene specific differences in tumor mutationalburden,PD-L1 expression, and outcomes from immunotherapy in non- small cell lung cancer. J ImmunoTher Cancer. 2021;9:e002891. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R13] [13].Qian C, Zhang Y, Cheng W, Zhang Q, Li M, Fang S. Case report: rechallenge with EGFR–TKIs after immunotherapy in EGFR–mutated non–small cell lung cancer with leptomeningeal metastasis. Front Oncol. 2022;12:957661. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R14] [14].Mazieres J, Drilon A, Lusque A, et al. Immune checkpoint inhibitors for patients with advanced lung cancer and oncogenic driver alterations: results from the IMMUNOTARGET registry. Ann Oncol. 2019;30:1321–8. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R15] [15].Yap TA, Parkes EE, Peng W, Moyers JT, Curran MA, Tawbi HA. Development of immunotherapy combination strategies in cancer. Cancer Discov. 2021;11:1368–97. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R16] [16].Paz-Ares LG, Ramalingam SS, Ciuleanu TE, et al. First-line nivolumab plus ipilimumab in advanced NSCLC: 4-year outcomes from the randomized, open-label, phase 3 checkmate 227 part 1 trial. J Thorac Oncol. 2022;17:289–308. [DOI] [PubMed] [Google Scholar]

[R17] [17].Schoenfeld JD, Giobbie-Hurder A, Ranasinghe S, et al. Durvalumab plus tremelimumab alone or in combination with low-dose or hypofractionated radiotherapy in metastatic non-small-cell lung cancer refractory to previous PD(L)-1 therapy: an open-label, multicentre, randomised, phase 2 trial. Lancet Oncol. 2022;23:279–91. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R18] [18].Lisberg A, Cummings A, Goldman JW, et al. A phase II study of pembrolizumab in EGFR-mutant, PD-L1+, tyrosine kinase inhibitor Naïve patients with advanced NSCLC. J Thorac Oncol. 2018;13:1138–45. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R19] [19].Garassino MC, Cho BC, Kim JH, et al. Durvalumab as third-line or later treatment for advanced non-small-cell lung cancer (ATLANTIC): an open-label, single-arm, phase 2 study. Lancet Oncol. 2018;19:521–36. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R20] [20].Chen N, Fang W, Lin Z, et al. KRAS mutation-induced upregulation of PD-L1 mediates immune escape in human lung adenocarcinoma. Cancer Immunol Immunother. 2017;66:1175–87. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R21] [21].Lee CK, Man J, Lord S, et al. Checkpoint inhibitors in metastatic EGFR-mutated non-small cell lung cancer-a meta-analysis. J Thorac Oncol. 2017;12:403–7. [DOI] [PubMed] [Google Scholar]

[R22] [22].Qiao M, Jiang T, Liu X, et al. Immune checkpoint inhibitors in EGFR-mutated NSCLC: dusk or dawn? J Thorac Oncol. 2021;16:1267–88. [DOI] [PubMed] [Google Scholar]

[R23] [23].Reck M, Rodríguez-Abreu D, Robinson AG, et al. Five-year outcomes with pembrolizumab versus chemotherapy for metastatic non-small-cell lung cancer with PD-L1 tumor proportion score ≥ 50. J Clin Oncol. 2021;39:2339–49. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R24] [24].West H, McCleod M, Hussein M, et al. Atezolizumab in combination with carboplatin plus nab-paclitaxel chemotherapy compared with chemotherapy alone as first-line treatment for metastatic non-squamous non-small-cell lung cancer (IMpower130): a multicenter, randomized, open-label, phase 3 trial. Lancet Oncol. 2019;20:924–37. [DOI] [PubMed] [Google Scholar]

[R25] [25].Socinski MA, Jotte RM, Cappuzzo F, et al. Atezolizumab for first-line treatment of metastatic nonsquamous NSCLC. N Engl J Med. 2018;378:2288–301. [DOI] [PubMed] [Google Scholar]

[R26] [26].Nogami N, Barlesi F, Socinski MA, et al. IMpower150 final exploratory analyses for atezolizumab plus bevacizumab and chemotherapy in key NSCLC patient subgroups with EGFR mutations or metastases in the liver or brain. J Thorac Oncol. 2022;17:309–23. [DOI] [PubMed] [Google Scholar]

[R27] [27].Roh W, Chen PL, Reuben A, et al. Integrated molecular analysis of tumor biopsies on sequential CTLA-4 and PD-1 blockade reveals markers of response and resistance. Sci Transl Med. 2017;9:eaah3560. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R28] [28].Bareche Y, Kelly D, Abbas-Aghababazadeh F, et al. Leveraging big data of immune checkpoint blockade response identifies novel potential targets. Ann Oncol. 2022;33:1304–17. [DOI] [PubMed] [Google Scholar]

PERMALINK

A retrospective analysis of immunotherapy in non-small cell lung cancer patients with targetable genetic mutations

Shiyu Zhang, MD

Jingwen Liu, MM

Jia Liu, MM

Kaihua Lu, MD

Abstract

1. Introduction