Role of genetic alterations on outcomes for pulmonary resection in oligometastatic non-small cell lung cancer.

Introduction
At diagnosis, approximately 50% of non-small cell lung cancer (NSCLC) patients present with metastatic disease (1,2). However, a subset of these individuals exhibits limited, rather than disseminated spread, categorized as oligometastatic disease (1). Hellman and Weichselbaum first described this state as an intermediate tumor burden level between those of locally and widely disseminated neoplastic processes (3). Thought-leaders in this space have argued that such patients can attain improved oncologic and survival outcomes with aggressive treatment of the primary tumor and all metastatic sites.
This concept has become the focus of several recent trials and retrospective studies (4-9). These investigations demonstrated that in oligometastatic (≤3 synchronous metastases) NSCLC, local therapy to all sites of disease with surgical and/or radiation therapy, described as comprehensive local consolidative therapy (cLCT), can improve progression-free survival (PFS) and overall survival (OS) compared to systemic therapy alone (4,10-18). Concurrently, the identification of genomic alterations in NSCLC, including EGFR and KRAS mutations, as well as ALK rearrangements, and others, has driven widespread testing for actionable mutations that may guide treatment strategies (17,19). While these advances have shown that the presence of specific genetic alterations has a clear impact on outcomes for lung cancer patients, minimal previous investigations characterized the influence of genomic variations, specifically EGFR, on outcomes for oligometastatic NSCLC patients undergoing cLCT (17,20-22). Furthermore, prior studies included patients who received heterogeneous cLCT modalities, leaving the outcomes for oligometastatic NSCLC patients with genomic alterations specifically undergoing lung resection as part of cLCT unclear. As such, we sought to ascertain the impact of genetic alterations on oncologic and survival outcomes in this patient population. We present this article in accordance with the REMARK reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1140/rc).

Methods

Study population
Following institutional review board approval (protocol 2023-0907; BASEC-reference number 2020-02566) with a waiver of informed consent, a retrospective review of databases prospectively maintained by the University of Texas MD Anderson Cancer Center and the University Hospital Zurich was performed. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Eligibility criteria included histological confirmation of stage IV NSCLC, treatment between January 1st, 1996 and December 31st, 2023, and oligometastatic disease (three or fewer synchronous metastatic lesions) at diagnosis (4-6,16). TNM-staging was defined according to the 8th Edition of the American Joint Commission on Cancer (23). All included patients underwent pulmonary resection as part of cLCT following multidisciplinary evaluation and treatment planning. Cohort identification was derived from using natural language processing (NLP) methods (10), and database/electronic-medical record (EMR) searches for those patients not flagged by NLP approaches. Collected variables were pulled from prospectively maintained databases and supplemented with EMR review.
Gathered datapoints included demographic, clinicopathologic, and treatment-related variables. Patient and tumor characteristics included sex, age at surgery, smoking history, primary tumor location, genomic alteration(s) of the primary tumor, and number of metastatic deposits. Patients were stratified by genomic alteration status: TP53, EGFR, KRAS, ALK, and no alterations. Additionally, individuals found to have 2 or more genomic alterations were distinctly categorized. The multiple genomic alterations discovered included ALK with TP53, EGFR with TP53, KRAS with TP53, as well as EGFR with KRAS. Gathered survival and oncologic outcomes included 30- and 90-day survival, PFS, and OS.

Variable and outcome definitions
As described in our previous publications, intrathoracic nodal disease, irrespective of the number of involved lymph nodes, was counted as a singular metastatic site, while multiple metastatic foci within an organ were considered as distinct sites of disease (10). Oligometastatic NSCLC was defined as three or fewer synchronous metastatic lesions, congruent with the definition utilized by prior research (4-6,16). Local consolidative therapy (LCT), may be defined as surgical resection or radiotherapy targeting the primary tumor and/or site(s) of metastases, which may or may not include all sites of disease. Comprehensive LCT, or cLCT is defined as the use of LCT modalities directed at the primary tumor and all metastatic sites, for complete consolidation (10).
OS was defined as the duration from diagnosis to patient death, while PFS measured the time from diagnosis to disease progression or recurrence. Patients alive and without disease progression or recurrence at the study’s end were censored in calculations on the last follow-up date.

Statistical analysis
Pearson chi-squared and Fisher’s exact test were utilized to analyze categorical variables, and Mann-Whitney U and Kruskal-Wallis tests were used to analyze continuous variables. OS and PFS were analyzed using the Kaplan-Meier method, while group differences were calculated using the log-rank test. Multivariable Cox regression analysis was used to determine the unadjusted and adjusted influences of clinicopathologic variables on OS and PFS. Variables included in the univariate Cox regression calculations were based on clinical importance and relevance, and included age, sex, smoking history, Eastern Cooperative Oncology Group (ECOG) performance status, genetic alteration, and neoadjuvant therapy. Statistical analyses were performed using IBM SPSS Statistics (Version 24). The P values reported are two-sided, with a P value <0.05 considered statistically significant.

Results

Patient population
In total, 87 patients met inclusion criteria, with 61 (70.1%) from MD Anderson Cancer Center and 26 (29.9%) from the University of Zurich (Table 1). EGFR mutations only (+EGFR) were identified in 12 (13.8%) individuals. Among the 75 patients with other alterations (-EGFR), identified genomic variations included TP53 (n=1/87, 1.2%), KRAS (n=11/87, 12.6%), and ALK (n=5/87, 5.7%), with no alterations found in 52 (59.8%) patients. Additionally, 6 individuals with multiple genomic alterations were found: ALK with TP53 (n=1/87, 1.2%), EGFR with TP53 (n=3/87, 3.4%), KRAS with TP53 (n=1/87, 1.2%), and EGFR with KRAS (n=1/87, 1.2%). The median age at diagnosis for the +EGFR group was 58.4 [interquartile range (IQR): 15.2] years, and 58.0 (IQR: 12.6, P=0.41) years for the −EGFR group. Both cohorts were fairly balanced by sex (+EGFR, n=7/12, 58.3%; −EGFR, n=37/75, 49.3%, P=0.56), and were predominantly White (+EGFR, n=10/12, 83.3%; −EGFR, n=64/75, 85.3%, P=0.92). Smoking history was more prevalent among individuals without EGFR mutations (n=64/75, 85.3%) compared to +EGFR patients (n=7/12, 58.3%, P=0.04). Regarding ECOG performance status, 8 (66.7%) of the +EGFR patients had an ECOG status of 0 or 1, and another 4 (33.3%) had a status of 2. Among all others, 46 (61.3%) had an ECOG status of 0 or 1, and 23 (30.7%, P=0.97) had an ECOG status of 2.
Clinical tumor status (cT) was cT1 in 4 (33.3%) +EGFR patients and 32 (42.7%) −EGFR patients, cT2 in 4 (33.3%) +EGFR patients and 27 (36.0%) −EGFR patients, cT3 in 1 (8.3%) +EGFR patients and 8 (10.7%) −EGFR patients, and cT4 in 3 (25.0%) +EGFR patients and 8 (10.7%, P=0.58, Table 2) −EGFR patients. As for clinical nodal status (cN), 6 (50.0%) of +EGFR and 44 (58.7%) −EGFR patients had cN1, 3 (25.0%) +EGFR patients and 12 (16.0%) −EGFR patients had cN2, 3 (25.0%) +EGFR patients and 16 (21.3%) −EGFR patients had cN2, and only 3 (4.0%, P=0.76) −EGFR patients had cN3. The most common location for metastatic spread for both groups was the brain (+EGFR, n=5/12, 41.7%; −EGFR, n=47/75, 62.7%, P=0.04). Other sites of metastatic involvement included bone, adrenal gland, contralateral lung, and pleura. Most individuals of both groups had one site of metastatic involvement (+EGFR, n=10/12, 83.3%; −EGFR, n=73/75, 97.3%, P=0.02).

Treatment details
Systemic therapy was commonly administered in the preoperative setting to all patients (Table 3). Among those without EGFR mutation, 34 (45.3%) received only chemotherapy, 4 (5.3%) received only targeted therapy, and 1 (1.3%) received chemotherapy and targeted therapy. As for +EGFR patients, 3 (25.0%) received only chemotherapy, 3 (25.0%) received only targeted therapy, and 2 (16.7%, P=0.03) received chemotherapy and targeted therapy. All patients underwent resection of their primary lung tumors as part of cLCT. Surgical intervention was utilized more often to consolidate distant disease among those without EGFR mutations (n=62/75, 82.7%) as compared to +EGFR patients (n=7/12, 58.3%), while rates of radiation therapy to metastatic sites were nearly equivalent between groups (+EGFR, n=8/12, 66.7%; −EGFR, n=56/75, 74.7%, P=0.60).
Thoracotomy was the most common surgical approach (+EGFR, n=6/12, 50.0%; −EGFR, n=58/75, 77.3%, P=0.07), and the majority of patients underwent lobectomy (+EGFR, n=8/12, 66.7%; −EGFR, n=59/75, 78.7%, P=0.56). R0 resection margins were achieved in most patients (+EGFR, n=10/12, 83.3%; −EGFR, n=68/75, 90.7%, P=0.65).
Among +EGFR patients, the most common pathologic tumor (pT) statuses were pT3 and pT4, each observed in 3 patients (25.0%). As for the other alteration status individuals, pT4 was the most commonly observed status (n=20/75, 26.7%, P=0.70, Table 3). Pathologic nodal (pN) status was most frequently pT0 for both groups (+EGFR, n=7/12, 58.3%; −EGFR, n=42/75, 56.0%, P=0.91), and most patients were found to have adenocarcinoma (+EGFR, n=11/12, 91.7%; −EGFR, n=62/75, 82.7%, P=0.39). While the majority of +EGFR individuals had moderately differentiated tumors (n=9/12, 75.0%%), other individuals predominantly had poorly differentiated tumors (n=47/75, 62.7%, P=0.01).

Oncologic and survival outcomes
Following lung resection, primary tumor site recurrence was uncommon among both groups (+EGFR, n=2/12, 16.7%; −EGFR, n=9/75, 12.0%, P=0.65, Table 4), as was recurrence at previously treated metastatic locations (+EGFR, n=1/12, 8.3%; −EGFR, n=20/75, 26.7%, P=0.17). Distant recurrence was common in both groups (+EGFR, n=6/12, 50.0%; −EGFR, n=41/75, 54.7%, P=0.73). Freedom from disease progression or recurrence was better among +EGFR patients (Figure 1). Median PFS in the +EGFR group was 43.1 months (95% CI: 22.14–64.07) and 19.1 months in the −EGFR group (95% CI: 16.46–31.74, P=0.58). Univariate Cox regression calculations showed no impact of EGFR alteration on PFS (HR =0.52, 95% CI: 0.20–1.34, P=0.18).
All 87 patients survived 30 days following pulmonary resection, and 90-day survival was achieved in 11 (91.7%) +EGFR patients and 74 (98.7%, P=0.13) other alteration status patients. Kaplan-Meier survival analyses revealed median OS of 88.9 months for +EGFR patients (95% CI: 48.46–121.28), and 30.8 months for –EGFR patients (95%CI: 12.25–49.36, P=0.20, Figure 2).
Notably, 3 (25.0%) +EGFR patients had bone metastasis. Among these individuals, 1 (8.3%) patient had only bone metastasis, 1 (8.3%) patient also had pleural metastasis, and 1 (8.3%) patient also had metastatic spread to the brain and adrenal gland. In terms of systemic therapy, the 2 (16.7%) individuals with bone and visceral metastases received targeted therapy in the neoadjuvant and adjuvant settings, whereas the individual with isolated bone metastasis only received adjuvant chemotherapy. The patient with only bone metastasis, as well as the patient with three total metastatic disease sites, experienced distant site recurrences following pulmonary resection. All of these individuals survived 30 and 90 days following primary tumor resection.
Multivariate Cox regression analysis was performed to determine predictors of PFS and OS (Table 5). Included covariates were those that had a significant impact on oncologic and survival outcomes in oligometastatic NSCLC. We found that the presence of EGFR alteration did not independently predict PFS (HR =0.68, 95% CI: 0.09–5.05, P=0.71) nor OS (HR =0.44, 95% CI: 0.05–7.23, P=0.62). However, cT3 and cT4 negatively impacted PFS (HR =2.99, 95% CI: 1.49–6.04, P=0.02) and OS (HR =3.48, 95% CI: 1.61–7.53, P=0.004).

Discussion
This present study represents a large, multicenter homogenous patient cohort with oligometastatic NSCLC who underwent cLCT, and to our knowledge, is the first to evaluate the impact of genomic alterations on outcomes following pulmonary resection in this patient population. Importantly, we unveiled a trend suggesting that the presence of an EGFR mutation may be associated with favorable OS when compared to other mutational statuses. Further, our findings corroborate recent reports highlighting the potential therapeutic benefits of cLCT in appropriately selected oligometastatic NSCLC patients (5,10,14,24).
Improvements in PFS and OS following LCT for oligometastatic NSCLC patients were first demonstrated in the landmark study by Gomez and colleagues (4,6). Pulmonary resection has also been shown to be a beneficial LCT modality in well-selected oligometastatic NSCLC patients, as shown by its positive impact on oncologic and survival outcomes (14). Subsequent studies revealed that LCT to the primary tumor and all sites of metastatic spread, termed cLCT, can improve OS when compared to patient populations who received LCT to only some disease sites (5). Our prior work also demonstrated substantial OS improvements with cLCT as compared to subcomprehensive LCT (21). Additionally, the presented results corroborate prior reports highlighting the feasibility of pulmonary resection as part of cLCT for oligometastatic NSCLC. Previously, we showed that pulmonary resection can enhance locoregional control while also improving oncologic and survival outcomes in a safe manner (10,25). Of key interest, recent clinical practice guidelines, building on the foundation of all previous work, recommend the consideration of LCT, including surgical consultation for all physiologically suitable patients with oligometastatic lung cancer (26).
The advent of next-generation sequencing for histologic subtypes and molecular drivers has led to the rise of targeted therapies, further transforming stage IV NSCLC care (27,28). Notably, 20% of non-Asian and 50% of Asian patients with NSCLC have tumors with EGFR mutations, and tyrosine kinase inhibitors (TKIs) have demonstrated a crucial role in improving these patients’ outcomes (29). Interestingly, compared to polymetastatic disease, oligometastatic NSCLC cases tend to express fewer genomic mutations, which highlights the significance of consolidating sites of disease (30). As such, first-line TKI, in conjunction with LCT, has been shown to provide survival benefits to oligometastatic NSCLC patients with EGFR mutations (22,31). Recently, we demonstrated that TKIs and cLCT confer the most favorable outcomes in this patient cohort, with significantly improved long-term survival (17). Moreover, brain metastases are commonly found in oligometastatic NSCLC patients with EGFR alteration, and better survival is seen in this cohort in comparison to similar patients without EGFR mutation, especially after receipt of TKI and LCT therapy (32).
In the present study, we considered the advances in systemic agents for advanced NSCLC in the contemporary era and sought to determine the impact of genomic alterations on outcomes in a large, multicenter cohort of oligometastatic NSCLC patients undergoing pulmonary resection. Importantly, and underscoring previous reports, we found that EGFR mutation of the primary tumor seemed to be associated with improved survival outcomes, though not statistically significant. Although the optimal timing of TKI and subsequent cLCT remains unclear, our results highlight the benefits of pursuing aggressive LCT strategies with pulmonary resection in carefully selected oligometastatic NSCLC patients with EGFR alteration. This is especially true in light of advances in both targeted therapies and LCT modalities. Ultimately, our findings underscore the importance of clinicians and surgeons to remain vigilant in their management of oligometastatic NSCLC patients, especially as our findings suggest that pulmonary resection may provide benefit to those with specific genetic alterations. Moreover, the Society of Thoracic Surgeons published guidelines that stress the importance of surgical decisions for stage IV NSCLC to be made in multidisciplinary settings, which should involve consideration of patient- and tumor-specific characteristics (26).
Further research is needed in order to uncover oncologic and survival differences among various genomic alterations. While targeted agent availability and efficacy can account for some observed distinctions in outcomes, the interplay among other variables, such as disease biology, patient demographics, comorbidities, and other potentially unknown factors must be investigated in order to further advance treatment paradigms for oligometastatic NSCLC. Discoveries in this space will inevitably lead to expanded therapeutic options for clinicians. Nevertheless, as a thoracic oncologic surgical community, we must maintain a commitment to aggressive management strategies while adapting to novel treatment modalities.
Despite its promising findings, this study has several limitations that must be acknowledged. The included patients were medically and surgically treated at two distinct institutions, which may have had different criteria for selecting surgical candidates. We only selected oligometastatic NSCLC patients who underwent pulmonary resection, with the inherent knowledge that these patients had baseline function and performance superior to those who did not receive lung resection. As such, these results may be less applicable to those who received alternate treatment modalities. Furthermore, the number of patients who underwent pulmonary resection, as well as the number of individuals with EGFR mutation, is part of a small sample size. Additionally, patients included in this study received treatment between 1996 and 2023. As mentioned, therapies for stage IV NSCLC have recently advanced, and their benefits would only be captured in those who received them.

Conclusions
This multicenter study evaluated the role of genomic alterations on the oncologic and survival outcomes of a homogenous cohort of oligometastatic NSCLC patients who underwent pulmonary resection as part of cLCT. Our findings support previous investigations regarding the therapeutic benefits of cLCT, along with the feasibility of performing pulmonary resection in well-selected individuals. Moreover, our results suggest that genomic alterations may play a role in the prognostication of oligometastatic NSCLC patients who undergo pulmonary resection as part of cLCT.

Supplementary
The article’s supplementary files as