Surgery is an effective treatment for non-small cell lung cancer (NSCLC), but the 5-year overall survival (OS) rates of patients with stage II and IIIA disease are only 65% and 41%, respectively (1). Even when the tumors in these patients have been radically resected, micrometastasis may exist before surgery and is considered to be the main factor causing postoperative local or distant recurrence. In addition to the elimination of micrometastases, preoperative systemic treatment could result in tumor shrinkage and decreased lymph node enlargement, therefore reducing the TNM stage and tumor burden and facilitating the surgical procedure. Therefore, the optimal neoadjuvant therapy should be to reduce tumor burden without delaying the scheduled operation and have fewer adverse effects. Studies have shown that the use of neoadjuvant chemotherapy can improve the OS of NSCLC patients (2). Although targeted therapy led by epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) and immunotherapy led by PD-1 inhibitors have been proven to be effective treatments in advanced NSCLC, the application of those reagents as neoadjuvant therapy for lung cancer other than chemotherapy is still at the exploration stage.

For the large group of patients with EGFR gene mutations, the administration of EGFR-TKIs is preferred (3, 4). Compared to the controversial molecular markers for the prediction of the efficacy of immunotherapy, the limited application of EGFR-TKIs in patients with EGFR wild-type NSCLC and the low abundance of EGFR mutations (5) are widely accepted. However, the design of previous neoadjuvant EGFR-TKI clinical trials did not distinguish between populations that were sensitive and those with wild-type mutations (6, 7). In more recent studies, the safety and efficacy of neoadjuvant EGFR-TKI therapy have been more focused on populations with EGFR-TKI-sensitive mutations (8, 9). In theory, sensitive mutations may improve the efficacy of neoadjuvant EGFR-TKIs, but there is currently little evidence to support this hypothesis.

In addition to the above clinical trial designs, which are based on changes in EGFR mutation status, a more detailed design taking clinical staging into consideration is needed. From 2009 to 2016, clinical trials tried to cover a broad spectrum of TNM stages, including patients from stage I to stage IV, and wild-type EGFR status (6, 10). Since 2016, the study designs have tended to focus on NSCLC patients with stage II and stage III disease (8, 9) with EGFR-sensitive mutations. In addition, a comparison of neoadjuvant EGFR-TKIs and chemotherapy (11, 12) suggests that neoadjuvant EGFR-TKIs can improve patient prognosis compared with chemotherapy. Nevertheless, the chemotherapy group in the study by Zhong et al. (11) was administered gemcitabine plus cisplatin, while in the study by Xiong et al. (12), cisplatin-based doublet chemotherapies including vinorelbine, gemcitabine, paclitaxel, docetaxel or pemetrexed were administered. In the cases of limited sample sizes and different chemotherapy combinations contributing as a confounding factor, the level of evidence for this conclusion needs to be improved by adding more results in future studies.

Though a series of phase II trials on neoadjuvant EGFR-TKIs for NSCLC have been reported, the safety and efficacy of neoadjuvant EGFR-TKIs, especially in subgroups of EGFR mutation status or TNM staging, remain unclear. Considering that these trials have great potential to change current neoadjuvant practice in lung cancer surgery, we performed a meta-analysis incorporating the results of the surgical rates, clinical responses, pathologic responses, toxicities, and prognoses to evaluate the safety and efficacy of neoadjuvant EGFR-TKI therapy.

We prospectively registered the protocol for this study in the International Prospective Register of Systematic Reviews (PROSPERO number: CRD42020187031). We reported the analysis by following the Meta-Analysis of Observational Studies in Epidemiology (MOOSE) standards ( Table S1 ) and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement.

Currently, neoadjuvant therapy based on chemotherapy has been proven to be effective (2, 15). The unsatisfactory overall response, adverse effects, and sometimes delay of surgery or inoperability, especially in the middle and late stages of NSCLC, require a more effective adjuvant treatment option. In this analysis, neoadjuvant EGFR-TKI therapy was shown to be a potential alternative for NSCLC patients with EGFR-TKI-sensitive mutations.

Compared with the overall response rates ranging from 50 to 70% depending on the combination (16) in neoadjuvant chemotherapy studies, the 48% ORR in the population with EGFR-TKI-sensitive mutations in this analysis seems to be acceptable. When considering the 45% stable disease rate in the population with EGFR-TKI-sensitive mutations including those with early-stage NSCLC, we hypothesize that the small EGFR-sensitive mutation tumors have relatively low abundances of EGFR mutations. Therefore, the improvement is not apparent. Of note, patients with advanced TNM stages may benefit more from neoadjuvant EGFR-TKIs, while early-stage patients may not benefit much, suggesting that there would be an optimal cutoff TNM stage to achieve a better neoadjuvant EGFR-TKI outcome. Furthermore, this study also suggests that the ORR can be significantly reduced with mixed wild-type mutation studies. Although we were unable to reanalyze the EGFR mutation status of subgroups of patients in some of the previous neoadjuvant chemotherapy studies, Zhong et al. (11) reported that the ORR for neoadjuvant erlotinib is better than that of gemcitabine plus cisplatin chemotherapy (54.1% versus 34.3%) in the EGFR-sensitive population. This finding suggests that this choice of chemotherapy in the EGFR-sensitive population as a neoadjuvant therapy could be inferior to the use of EGFR-TKIs. Of course, more definite decision making depends on more evidence from clinical trials in the future.

Another indicator of whether a drug is suitable for neoadjuvant therapy is the occurrence and level of the AEs. Although preoperative chemotherapy has advantages, its toxicity and side effects cannot be ignored. In addition to affecting liver and kidney functions, chemotherapy drugs also present toxicities in the cardiovascular and nervous systems. Although the occurrence of side effects is closely related to the dose and combination of chemotherapy drugs, in general, EGFR-TKIs have fewer side effects. Similar to the findings of previous EGFR-TKI studies (17), the most common side effects in neoadjuvant EGFR-TKI studies were rash and diarrhea. Although more than half of the patients had grade 1-2 AEs, fortunately, only a small number of patients had grade 3-4 AEs, possibly avoiding the accumulating toxicity from the long-term use of EGFR-TKIs in previous clinical trials. This confers one of the essential factors to ensure that the surgery is performed as scheduled.

There are many limitations in current neoadjuvant EGFR-TKI studies. First, this study has not been included results from ongoing clinical trials on the neoadjuvant therapy with next-generation EGFR-TKIs, such as afatinib (NCT04201756) or osimertinib (NCT03433469). Second, most of the current clinical trials based on patients with EGFR-TKI-sensitive mutations are from China, and more evidence from the Caucasian population is needed. The current studies inconsistently reported the effect of neoadjuvant EGFR-TKIs on survival (only 2 out of 7 studies), so it is not rational to perform a pooled analysis for this outcome. Further studies on this limitation are warranted. It is unknown whether the combination of neoadjuvant EGFR-TKIs and chemotherapy or immunotherapy achieves a better response rate and prolongs NSCLC survival. Finally, for the intrinsic few studies have been reported in and the methods of controlling were different among studies, the heterogeneity cannot be ignored. An updated meta-analysis is needed in the future.

We first provided pooled estimates of the surgical rate, response rate, and drug toxicity rate in patients receiving neoadjuvant EGFR-TKIs. Our analysis revealed that EGFR-TKIs are a promising neoadjuvant option for NSCLC patients with EGFR-TKI-sensitive mutations. Potential factors that affect these estimates were also investigated. Our findings indicate that neoadjuvant EGFR-TKIs could be more effective in NSCLC patients with EGFR-TKI-sensitive mutations than in those with wild-type EGFR.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

Conceptualization: KC and XS. Data collection and literature screening: XD, JZ, XL, and ZN. Data curation: XS, XD, and HW. Data analysis: XS. Funding acquisition: XS and KC. Methodology: XS and XD. Writing—original draft: DL, JZ and XS. Writing—review and editing: DL, ZN, XD, and XL. All authors contributed to the article and approved the submitted version.

The work is partially supported by the Research Initiative Fund of Southern Hospital 2018 (C1051325) and the National Natural Science Foundation of China (81902319), the Major Science and Technology Planning Project of Guangdong Province (2017B020226005), and the Science and Technology Program of Guangzhou City (201903010003).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.