Clinicopathologic outcomes of preoperative targeted therapy in patients with clinical stage I-III non-small cell lung cancer

Abstract

Objective:

Targeted therapy improves outcomes in patients with advanced-stage non-small cell lung cancer (NSCLC) and in the adjuvant setting, but data on its use before surgery are limited. We sought to investigate the safety and feasibility of preoperative targeted therapy in patients with operable NSCLC.

Methods:

We retrospectively reviewed 51 patients with clinical stage I-III NSCLC who received targeted therapy, alone or in combination with chemotherapy, before surgical resection with curative intent, treated from 2004 to 2021. The primary outcome was the safety and feasibility of preoperative targeted therapy; secondary outcomes included objective response rate, major pathologic response (defined as ≤10% viable tumor) rate, recurrence-free survival (RFS), and overall survival.

Results:

Of the 51 patients included, 46 had an activating EGFR alteration and 5 had an ALK fusion. Overall, 37 of 46 evaluable patients experienced at least 1 adverse event before surgery; however, only 3 patients experienced a grade 3 or 4 event. The objective response rate was 38% (17/45) for all evaluable patients and 44% (14/32) for patients with clinical stage II or III disease. The major pathologic response rate was 20% (9/44); 2 patients had a complete pathologic response. Median RFS was 3.8 years (95% CI, 2.8 to not reached). Targeted therapy alone was associated with better RFS than combination therapy (p=0.009) in patients with clinical stage II or III disease.

Conclusions:

Preoperative targeted therapy was well tolerated and associated with good outcomes, with or without induction chemotherapy. In addition, radiographic response and pathologic response were strongly correlated.

Graphical Abstract

Introduction

Non-small cell lung cancer (NSCLC) accounts for approximately 80% of all lung cancers, with surgical resection the optimal treatment for early-stage disease. However, even with a complete resection, an estimated 20-60% of patients develop recurrence, ranging from 11% in pT1aN0 tumors to over 60% of clinical stage III-N2 with induction therapy.^1–5 Given the propensity for recurrence for node-positive or larger tumors, the National Comprehensive Cancer Network (NCCN) guidelines recommend adjuvant chemotherapy for patients with pathologic stage II or IIIA disease.^{6, 7} The benefit in 5-year overall survival (OS) with adjuvant platinum-based regimens is only 5.4%, and this comes at the cost of significant toxicity in many patients.^{7, 8} Comparisons of neoadjuvant and adjuvant chemotherapy with or without radiation for operable NSCLC have revealed similar outcomes.^{2, 9} Because delivery of adjuvant systemic therapy remains challenging for some patients following surgical resection, the NCCN guidelines recommend considering neoadjuvant treatment using platinum-doublet chemotherapy with or without immunotherapy in any patient that would likely be considered for adjuvant therapy.^{10, 11}

Next-generation sequencing has been used to identify specific oncogenic drivers in NSCLC, and the subsequent development of targeted therapies has revolutionized lung cancer treatment paradigms.^{12, 13} Tyrosine kinase inhibitor (TKI)–based therapies are now a first-line option for advanced-stage NSCLC with targetable oncogenic drivers.^{14, 15} Recently, the ADJUVANT-CTONG and ADAURA trials demonstrated that the use of adjuvant epidermal growth factor receptor (EGFR) TKIs plus chemotherapy resulted in significantly better disease-free survival in patients with resectable NSCLC.^16–18 Given this success for both early- and advanced-stage disease, interest in integrating targeted therapies to neoadjuvant treatment regimens has grown. Although a few studies have explored the use of neoadjuvant TKIs in operable NSCLC, the available data remains limited.^{16, 19–22}

To help address this knowledge gap, we retrospectively reviewed patients who received planned, matched targeted therapy before surgical resection for NSCLC. The primary aim of the study was to assess the safety and feasibility of preoperative TKIs; secondary aims were to determine pathologic responses and clinical outcomes.

Methods

Patient Cohort

This study was approved by the institutional review board at Memorial Sloan Kettering Cancer Center, and all patients gave informed consent (MSK IRB#16-1395; approved 8/26/2016). We retrospectively reviewed patients with NSCLC who were treated from January 2004 to April 2021 with targeted therapy before surgical resection with curative intent. At the time of initial evaluation at our institution, all patients were planned for surgical resection following completion of their preoperative therapy. However, some patients were initially begun on systemic therapy at an outside institution which may account for some variation in treatment courses.

From an initial 156 patients who received targeted therapy before surgical resection, 51 patients met the inclusion criteria: clinical stage I-III lung adenocarcinoma and curative resection after treatment with targeted therapy either alone or with cytotoxic chemotherapy. Patients with a known mutation matched to an FDA-approved targeted therapy were included. Clinical stage I patients are not routinely recommended for neoadjuvant therapy but were treated as part of a clinical trial and were included for safety and feasibility evaluation.²³ Choice of targeted therapy alone or combination therapy was decided based upon clinical trial criteria or at physician discretion. Patients with combination therapy tended to have more advanced clinical disease and received cytotoxic chemotherapy as standard of care with addition of targeted therapy drugs. However, over time some patients were transitioned to targeted therapy alone. Exclusion criteria included unknown mutation status (n=8), wild-type or nonsensitizing mutations for the targeted therapy drug in question (n=45), planned definitive therapy with prolonged TKI course (n=4), and clinical stage IV NSCLC (n=48) treated with resection after downstaging or for oligometastatic or oligoprogressive disease (see CONSORT diagram, Supplemental Figure 1).

Clinical characteristics, preoperative imaging (including computed tomography [CT] and positron emission tomography), and pathologic reports were reviewed. The presence of known targeted alterations was confirmed on the basis of tissue biopsy specimens using either select sequencing with EGFR mutations or broad-panel next-generation sequencing with MSK-IMPACT.²⁴ Select testing for EGFR mutations was performed with exon 19 deletion analysis, exon 21 L858R mutation analysis, or Idylla EGFR testing.²⁵ ALK rearrangements were confirmed by immunohistochemistry for D5F3 clone or by fluorescence in situ hybridization analysis and confirmed with MSK-IMPACT testing in 4 of 5 patients.^{26, 27}

Response Evaluation

CT scans were performed to assess treatment response. Radiologic tumor response was assessed retrospectively using Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 by a thoracic radiologist (C.C.L.). Responses were classified as progressive disease (diameter increase of ≥20% or development of new lesions), stable disease (diameter change between 20% increase and 30% decrease), partial response (PR; diameter decrease of ≥30%), and complete response (CR; no residual disease on imaging).²⁸

Study Outcomes

The primary outcome was the safety and feasibility of preoperative TKIs. Adverse events (AEs) related to preoperative therapy were assessed using National Cancer Institute Common Terminology Criteria for Adverse Events versions 3 and 4.¹³ In-hospital complications, 30-day and 90-day complications, and 90-day readmissions were also assessed. All in-hospital and surgical complications were graded on the basis of the Clavien-Dindo classification system, with grade 3 and 4 AEs representing major morbidity.²⁹ Secondary outcomes included major pathologic response (MPR; defined as ≤10% viable tumor) and pathologic complete response (PCR; defined as no remaining viable tumor), which were assessed after resection.³⁰ Given changes in pathologic reporting, MPR information was available for 44 of the 51 patients. Clinical outcomes included objective response rate (ORR), pathologic response, recurrence-free survival (RFS), and OS. ORR was defined as the percentage of patients who had either a PR or a CR (RECIST version 1.1) after preoperative TKI treatment.²⁸ OS and RFS started from the date of surgery to the date of event or the last follow up. When features associated with RFS and OS were assessed, patients with clinical stage I NSCLC were excluded, as they are known to have a better prognosis.

Follow-up

Follow-up was performed in accordance with NCCN guidelines.^{1, 8} Recurrences were distinguished from metachronous tumors using the Martini and Melamed criteria.³¹ When available, genomic and pathologic data were used to confirm relatedness.³² The timing and location of all recurrences were recorded. RFS was defined as the length of time from surgery to the day of the first recurrence. Patients without recurrence were censored at the date of the last follow-up. OS was defined as the length of time from surgery to the day of death, with all other patients censored at the date of the last follow-up. Locoregional recurrence (LR) was defined as recurrence at the primary tumor site, ipsilateral hilar lymph nodes, ipsilateral mediastinal lymph nodes, or ipsilateral supraclavicular lymph nodes. Distant metastasis was defined as recurrence at any other site.

Statistical Analysis

We summarized targeted therapy AEs, radiographic response, ORR, MPR, PCR, and postoperative complications as frequency and percentage. As some patients had multiple AEs, the Rao-Scott Chi-squared test was used to compare AE grade between patients receiving targeted therapy alone and those receiving combination therapy. The association between radiographic response and MPR was examined using the Chi-squared test. The Kaplan-Meier method was used to examine OS and RFS. Differences in survival outcomes between different groups were examined using the log-rank test. We also estimated the restricted mean survival time (RMST) at 3 years from the surgery, along with the 95% confidence interval (Supplemental Table 1 and 2). The estimated RMST was interpreted as the average time patients were event free when they were followed up to 3 years. Because of a limited number of events, multivariable analysis was not attempted. The Wilcoxon rank-sum test or Fisher’s exact test was used to examine associations between neoadjuvant treatment type and other variables. Statistical analyses were two-sided, and p<0.05 was considered statistically significant. All statistical analyses were performed using R version 4.1 (R Foundation for Statistical Computing).

Results

Demographic Characteristics

In total, 51 patients met inclusion criteria, of which 90% (46/51) had an EGFR alteration and 10% (5/51) had an ALK fusion. Of the overall cohort, 35 were women (35/51 [69%]), and 27 (27/51 [53%]) were never-smokers. The median age at surgery was 64 years (interquartile range, 55-72 years). The majority of patients (46/51 [90%]) underwent lobectomy, and all patients had an R0 resection. Surgical approach evolved over the course of the study, with a majority of earlier patients undergoing open resection (27/51 [53%]). Four patients required conversion from a minimally invasive approach to an open procedure (4/24 [17%]). Forty-four patients received adjuvant systemic therapy (44/51 [86%]), and 37 patients (37/51 [73%]) continued targeted therapy (Table 1).

Table 1.

Patient characteristics (N=51)

Variable	All Patients (n=51)
Sex
Male	16 (31)
Female	35 (69)
Age, years	64 (55-71.5)
Smoking history
Ever	24 (47)
Never	27 (53)
FEV1, %	104 (88-107)
DLCO, %	94 (82-103)
Targeted mutation type
EGFR exon 19 deletion	27 (53)
EGFR L858R	19 (37)
ALK fusion	5 (10)
cTNM stage
I	14 (27)
II	12 (24)
III	25 (49)
Surgical approach
Open	31 (61)
Minimally invasive or VATS	20 (39)
Surgical procedure
Wedge resection	1 (2)
Lobectomy	46 (90)
Bilobectomy	1 (2)
Pneumonectomy	3 (6)
ypTNM stage
0	2 (4)
I	21 (41)
II	9 (18)
III	18 (35)
IV	1 (2)
Adjuvant systemic therapy
Any adjuvant systemic therapy	44 (86)
Targeted therapy (+/− chemotherapy)	37 (73)

Data are no. of patients (%) or median (interquartile range). cTNM, clinical tumor, node, metastasis; DLCO, diffusing capacity for carbon monoxide; FEV1, forced expiratory volume in 1 second; VATS, video-assisted thoracic surgery; ypTNM, pathologic tumor, node, metastasis after therapy.

Preoperative Treatment

Given the retrospective nature of this study, no standardized treatment duration was prescribed before surgical resection. The median duration of targeted therapy was 61 days (range, 14-176 days). The most used drug was gefitinib (n=18), followed by erlotinib (n=17) and osimertinib (n=10). The choice of targeted therapy evolved over time: with osimertinib first administered in this cohort in 2018. Four patients with an ALK fusion received alectinib, and 1 received crizotinib. In total, 16 patients (16/51 [31%]) received preoperative combination therapy consisting of targeted therapy plus other systemic chemotherapy drugs. No patients received preoperative immunotherapy or >1 targeted therapy (Table 1).

Preoperative AEs

Five patients (5/51 [10%]) received neoadjuvant targeted therapy at another institution, and data on preoperative treatment details and AEs were not available for them. Thirty-seven patients (37/46 [80%]) had at least 1 reported AE before surgical resection, for a total of 81 reported AEs (Table 2). The most common toxicities were skin toxicity and gastrointestinal symptoms, which occurred in 48% (22/46) and 46% (21/46) of patients, respectively. Only 3 patients (3/46 [7%]) experienced a grade 3 or 4 event, which we classified as a major morbidity. However, as a result of drug toxicity, 9 patients (9/46 [20%]) required dose reduction or interruption of their targeted therapy, with an eventual 5 patients (5/46 [11%]) discontinuing targeted therapy early. No patients had an AE resulting in a surgical delay.

Table 2.

Adverse events and postoperative complications

Event or Complication	Grade 1 or 2	Grade 3 or 4
Preoperative adverse event (n=46)
Skin toxicity	22 (48)	1 (2)
Gastrointestinal symptoms	21 (46)	2 (4)
Fatigue	11 (24)	0 (0)
Mucositis or stomatitis	8 (17)	0 (0)
Cough	4 (9)	0 (0)
Muscle discomfort	4 (9)	0 (0)
Dyspnea	2 (4)	0 (0)
Arthralgia	2 (4)	0 (0)
Epistaxis	1 (2)	0 (0)
Edema	1 (2)	0 (0)
Electrolyte disturbance	1 (2)	0 (0)
Paronychia	1 (2)	0 (0)
Postoperative or in-hospital complication (n=51)
Airleak	5 (10)	0 (0)
Chylothorax	2 (4)	0 (0)
Anemia	2 (4)	0 (0)
Atelectasis	0 (0)	1 (2)
Atrial fibrillation	1 (2)	0 (0)
Wound infection	2 (4)	0 (0)

Data are no. of patients. (%).

Postoperative Complications

Nine patients (9/51 [18%]) experienced a postoperative in-hospital complication, with 1 patient (1/51 [2%]) experiencing grade 3 atelectasis requiring a return to the operating room for right middle lobe torsion. The most common postoperative complication was air leak (n=5); other AEs included chylothorax, atrial fibrillation, infection, and anemia requiring blood transfusion (Table 2). Overall, 13 patients (13/51 [25%]) experienced a complication within 90 days of surgery. Ninety-day postoperative mortality was 0%.

Radiographic Response

Radiographic response is commonly used as a surrogate for neoadjuvant treatment effect. In our study, pretreatment and presurgery CT scans to assess treatment response were available for review for 45 of 51 patients (88%). Seventeen patients experienced a PR, and no patient had a CR, resulting in an ORR of 38%. For patients with clinical stage I NSCLC the ORR was 23% (3/13) whereas for patients with clinical stage II or III NSCLC, the ORR was 44% (14/32). Patients who received only targeted therapy had a higher ORR than patients who received combination therapy (44% [14/32] for only targeted therapy vs. 23% [3/13] for combination therapy) although it was not statistically significant. Patients with a response to therapy tended to have higher stage disease; only 3 of 13 patients (23%) with clinical stage I NSCLC had a PR, although many patients with clinical stage I NSCLC were treated with short durations of neoadjuvant therapy, which is not traditionally used in assessment of radiographic response. Additionally, when patients were stratified by mutation type, those with an ALK fusion had the best rate of radiographic response, with an ORR of 80% (4/5); in contrast, the ORR for patients with EGFR exon 19 deletions was 38% (9/24) and for patients with EGFR L858R mutations was 25% (4/16) (overall p=0.08). Overall, the median maximum percentage change in tumor diameter was a 26% decrease (range, 8% increase to 57% decrease) (Figure 1).

Figure 1.

Waterfall plot of the maximum change in tumor size in patients after preoperative targeted therapy, by mutation type. Each bar represents the best response in an individual patient. Dashed lines represent divisions between Response Evaluation Criteria in Solid Tumors categories.

Pathologic Response

Pathologic response to neoadjuvant treatment is an important prognostic factor in NSCLC. In our cohort, 9 (20%) patients had an MPR; this includes 2 patients with a PCR. Of the 9 patients with an MPR, 8 received only targeted therapy, and 1 received combination therapy. Both patients with PCR received only targeted therapy and, at the end of follow-up, were alive without recurrence. Additionally, 31% of patients (16/51) had pathologic downstaging. A complete description of pathologic downstaging is provided in Supplemental Table 3. There was a strong association between pathologic response and radiographic response: 53% of patients (7/13) with a PR had an MPR versus 8% of patients (2/26) with stable disease (p=0.003); however, patients with a PR on imaging tended to have larger tumors and were treated for longer durations.

Clinical Outcomes

The median follow-up from surgery was 9.8 years (95% CI 6 to 9.2 years). At last follow-up, 20 patients had disease recurrence, and 19 patients had died. Median OS was 9.2 years (95% CI, 4.7 to not reached), and median RFS was 3.8 years (95% CI, 2.8 to not reached) (Figure 2). Of the patients who had disease recurrence, the median time to first recurrence was 1.73 years (interquartile range, 0.76-2.83 years). Seventeen patients developed a distant metastasis of which the most common sites were brain (10/20 [50%]), lung (8/20 [40%]), and bone (6/20 [30%]). Osimertinib has been shown to be associated with better efficacy against central nervous system metastases.^33–35 Of the 10 patients treated with osimertinib in this study, no patient experienced a brain metastasis, and only 1 developed a recurrence (bone metastasis), with a median follow-up of 1.15 years.

Figure 2.

Kaplan-Meier curves of (A) recurrence-free survival and (B) overall survival for all patients (95% CI).

In a univariable analysis that investigated features associated with RFS and OS, preoperative therapy type (targeted alone or combination therapy), pathologic stage, and pathologic T stage were associated with RFS, whereas only pathologic T stage was associated with OS (Table 3). Patients who received only targeted therapy had better RFS than patients who received combination therapy, both in the entire cohort and among only patients with clinical stage II or III NSCLC (Figure 3). However, patients who underwent preoperative combination therapy, in comparison to targeted therapy alone, more often had higher pathologic stage (pathologic stage 2-4, 81% vs. 52%; p=0.09) and larger tumors on final pathology (pathologic T stage, 56% vs. 24%; p=0.09), which may explain the worse RFS in this group.

Table 3.

Univariable survival analysis in patients with clinical stage II or III disease

	RFS		OS
Variable	3-Year RFS (95% CI), %	P	3-Year OS (95% CI), %	P
Sex		0.58		0.83
Male	41 (18-95)		78 (55-100)
Female	45 (26-78)		69 (49-96)
Age		0.54		0.16
<65 years	44 (24-82)		85 (68-100)
≥65 years	42 (20-85)		56 (32-96)
Targeted genomic alteration		0.51		0.34
EGFR exon 19 deletion	50 (29-86)		71 (50-100)
EGFR L858R mutation	34 (14-82)		71 (48-100)
ALK fusion	NA		100 (100-100)
Preoperative treatment type		0.009		0.27
Targeted therapy only	67 (44-100)		92 (79-100)
Combination (targeted therapy + chemotherapy)	27 (12-62)		60 (40-91)
ypTNM stage		0.001		0.23
0-1	71 (45-100)		86 (63-100)
2-4	28 (13-62)		65 (45-93)
pT stage		0.003		0.002
0-1	65 (43-97)		94 (83-100)
2-4	18 (5-61)		45 (23-89)
Surgical procedure		0.141		0.30
Lobectomy or wedge resection	44 (27-72)		73 (56-95)
Bilobectomy or pneumonectomy	33 (7-100)		67 (30-100)
Radiographic response (RECIST version 1.1)		0.19		0.71
Partial response	61 (32-100)		83 (58-100)
Stable disease	37 (19-72)		66 (45-96)
Pathologic response		0.11		0.25
Major pathologic response
Yes	100 (100-100)		100 (100-100)
No	32 (16-65)		58 (38-88)

P values were derived from log-rank tests assessing differences between survival curves by groups. Bold indicates statistically significant values. NA, not achieved; OS, overall survival; RECIST, Response Evaluation Criteria in Solid Tumors; RFS, recurrence-free survival; ypTNM, pathologic tumor, node, metastasis after therapy.

Figure 3.

Kaplan-Meier curves of recurrence-free survival for (A) all patients with preoperative targeted therapy and (B) only patients with clinical stage II or III disease. For each plot, patients were stratified by type of preoperative therapy (targeted therapy only and combination therapy) (95% CI).

Discussion

Targeted therapies are associated with longer survival among patients with advanced-stage NSCLC and when given as adjuvant therapy after complete surgical resection.^{8, 12, 16, 33} In contrast, only a few studies, primarily in East Asian populations,^{36, 37} have examined the use of TKIs in the neoadjuvant setting.^{20, 21, 38, 39} Our study is the largest retrospective study, to date, to investigate preoperative TKI use in patients with activating EGFR mutations and ALK fusions. We found that preoperative TKI use was well tolerated, with very few serious AEs and good outcomes. Our results support the potential use of targeted therapy in the neoadjuvant setting.

We assessed the safety and feasibility of preoperative TKI use as the primary endpoint of our study. Although 80% of patients in our study experienced at least 1 preoperative AE, nearly all were grade 1 or 2 events, and there was a low rate (11%) of treatment discontinuation with no surgical delays. In contrast, neoadjuvant platinum-based chemotherapies have been associated with substantial AEs in 50% of patients.^{40, 41} In the ADAURA trial, grade 3 or 4 AEs occurred in 20% of patients who received adjuvant osimertinib, with 11% requiring treatment discontinuation.³³ Neoadjuvant regimens are shorter in duration than adjuvant regimens but have also been associated with frequent, albeit mild, AEs. In the study by Zhang et al., 86% of patients treated with neoadjuvant gefitinib experienced at least 1 AE, although there were no grade 3 or 4 events.²⁰ Concerns regarding neoadjuvant therapy often involve surgical delays or the potential for more-difficult procedures. However, in our study, the use of preoperative targeted therapy was associated with a 90-day postoperative complication rate of 25% and 0 deaths, which are lower than the reported rates after neoadjuvant immunotherapy or chemotherapy alone.^{10, 40, 41} Additionally, although 4 patients (representing 16% of minimally invasive cases) required conversion, all of these occurred from 2004 to 2007. One could hypothesize early conversions may be related to surgeon experience as the percentage of minimally invasive resections increased over time, from 32% in the first 4 years to 73% in the last 4 years of the study. Overall, our safety results, which are comparable to those of previous reports and include few serious preoperative AEs or postoperative complications, are of particular importance, as 31% of patients (16/51) received a combination of targeted therapy and platinum-based chemotherapy, which is associated with higher rates of toxicity.^{14, 42–45}

Radiographic response is used as a surrogate marker for the efficacy of neoadjuvant therapy, although its relationship with clinical outcomes remains unclear.^{20, 21, 46–49} In our study, the ORR was 38%, which is comparable to rates in previous reports after neoadjuvant TKI treatment, which range from 42.1% to 54.5%.^{20, 21, 38, 39} Most studies that investigated neoadjuvant targeted therapy examined only patients with clinical stage II or III disease, whereas our study also included patients with clinical stage I disease (n=14), who received targeted therapy for a median of 21 days (range, 14-38 days). The ORR was 23% for patients with clinical stage I disease and 44% for patients with clinical stage II or III disease, which may be attributable to the shorter treatment duration and smaller tumors in clinical stage I patients. Interestingly, we found differences in ORR among mutation types and specific TKIs. For example, all 4 patients treated with alectinib had a PR; in contrast, of the patients with EGFR mutations, 38% of patients with exon 19 deletions and 25% of patients with L858R mutations had a PR.

Pathologic response has been linked to patient outcomes in NSCLC and is not subject to factors such as tumor stage or timing of imaging, which can affect radiographic response.^{30, 50–52} In the CheckMate 816 trial, which evaluated neoadjuvant nivolumab plus chemotherapy versus chemotherapy alone, the MPR was 36.9% in the nivolumab group versus 8.9% in the chemotherapy only group, with respective PCR of 24.0% and 2.2%.¹⁰ In our cohort, preoperative targeted therapy was associated with better rates of pathologic response than previously reported chemotherapy alone. Among patients who received preoperative targeted therapy, 20% (9/44) had an MPR while 2 patients had a PCR. This is particularly important because patients with targetable EGFR and ALK alterations are typically ineligible for immunotherapy-based trials. Furthermore, these findings highlight the need for further research on the use of neoadjuvant therapies in these patients. The MPR in our study is consistent with those in previous reports of neoadjuvant targeted therapy, which range from 10% to 24%.^{20, 21}

With a median follow-up of over 5 years, our results demonstrate good RFS and OS (median RFS, 3.8 years; median OS, 9.2 years). Analysis of patients with clinical stage II or III NSCLC revealed that combination therapy was associated with worse RFS, as were pathologic stage and tumor size. Interestingly, although MPR has previously been reported to be associated with better disease-free survival,^{10, 20} MPR was not associated with better RFS or OS in our study. This may be explained by a lack of statistical power and patient heterogeneity. However, whether the better disease-free survival observed with the use of TKIs can serve as surrogate for OS or whether it overestimates the true survival benefit by suppressing oncogene-driven cancer without improving OS once treatment is removed remains a matter of debate.⁵³

As this is a single-center, retrospective study, the generalizability of our findings may be limited. In addition, the study cohort spans 15 years, with variation in both the specific TKIs and the treatment regimens used, which results in a heterogenous patient population. The use of early generation TKIs prior to 2018 may underestimate the potential survival benefit with current TKI options, though, controversy remains regarding the OS benefit of TKI therapy.⁵³ Included in this study is a subset of patients with early-stage disease who underwent shorter neoadjuvant treatment courses and therefore may not have had adequate time to develop a radiographic response.²³ Despite this, we can infer the general safety and feasibility of preoperative targeted therapy. Other prognostic factors, such as predominant tumor histologic subtype and spread through air spaces, were not fully accounted for, given changes in pathologic review over time. Finally, while all patients received targeted therapy and underwent surgery, we did not assess the relative survival benefits compared with other neoadjuvant strategies or surgery alone. We also did not evaluate patients who underwent neoadjuvant targeted therapy but did not proceed to surgery. To help address these questions, 2 large prospective trials, NeoADAURA and the Lung Cancer Mutation Consortium LEADER screening trial, are currently investigating the role of neoadjuvant TKI therapy in patients with tumors with preoperatively identified oncogenic driver alterations.^{19, 54}

Conclusions

Preoperative TKI targeted therapy was well tolerated in patients with clinical stage I-III NSCLC with actionable EGFR and ALK alterations (Figure 4). In addition, radiographic response strongly correlated with pathologic response, and preoperative targeted therapy, even when used alone, was associated with good RFS and OS.

Figure 4.

The methods, results, and implications of this study. Preoperative targeted therapy was well tolerated, with few serious adverse events and good clinical outcomes. Targeted therapy alone was associated with better recurrence-free survival than combination therapy.

Central Picture.

Preoperative targeted therapy is safe and associated with good clinical outcomes.

Central Message:

Preoperative targeted therapy using EGFR and ALK tyrosine kinase inhibitors, with or without concurrent chemotherapy, for NSCLC is safe, with few serious adverse events and good clinical outcomes.

Perspective Statement:

Preoperative targeted therapy was well tolerated, with good clinical outcomes, in patients with stage I-III NSCLC with oncogenic EGFR and ALK alterations. TKIs, with or without standard chemotherapy, were associated with few treatment-limiting toxicities, and radiographic response correlated with pathologic response. Future studies are needed to evaluate surrogate endpoints and outcomes.

Abbreviations:

AE

adverse event

CR

complete response

LR

locoregional recurrence

MPR

major pathologic response

NCCN

National Comprehensive Cancer Network

NSCLC

non-small cell lung cancer

ORR

objective response rate

OS

overall survival

PCR

pathologic complete response

PR

partial response

RECIST

Response Evaluation Criteria in Solid Tumors

RFS

recurrence-free survival

TKI

tyrosine kinase inhibitor

Biographies