Lung cancer has the second-highest incidence rate and the highest mortality among all cancers worldwide (1), with non-small cell lung cancer (NSCLC) accounting for approximately 85% of all lung cancer cases (2). Radiotherapy has become one of the primary treatment modalities for NSCLC. The main mechanism of radiotherapy is that ionizing radiation directly or indirectly (through reactive oxygen species) damages Deoxyribonucleic acid, triggering a series of events that may lead to cell death (3). However, due to the non selective killing effect of radiotherapy on cells (3), if it extends too much outward on the base of gross tumor volume (GTV), it will increase the irradiation volume of healthy tissue, thereby increasing the risk of adverse reactions. Therefore, in radiotherapy, the precision of the target volume delineation greatly affects the safety and effectiveness of this treatment, especially the limits of the clinical target volume (CTV) (4). Giraud et al. (4) quantitatively evaluated the microscopic tumor extension (ME) of NSCLC by conducting a pathological analysis of specimens, after surgery from patients with NSCLC who underwent direct surgery (DS). Their findings suggest that the CTV of lung adenocarcinoma (ADC) and squamous cell carcinoma (SCC) should be extended outward by 8 mm and 6 mm, respectively, based on the GTV (4).

Recently, with the advancement of cancer research, preoperative neoadjuvant therapy has become the standard treatment method for operable NSCLC. Surgery after neoadjuvant chemotherapy (NAC) provides better survival benefits than does direct surgical treatment (5–9). Additionally, with the rise in immunotherapy, several recent studies have demonstrated that patients receiving neoadjuvant chemotherapy combined with immunotherapy (NACI) have a more significant survival benefit than that of those receiving NAC (10–12). However, some patients undergoing neoadjuvant therapy do not undergo surgical treatment for various reasons and choose to undergo radiotherapy. For inoperable NSCLC, sequential chemotherapy-radiotherapy has better therapeutic effects than radiotherapy alone (13, 14). While the optimal timing and strategy for immunotherapy intervention require further investigation and validation, several studies have preliminarily confirmed the feasibility of the early immunotherapy application (15, 16). Nevertheless, no studies have so far defined specific margins of GTV expansion for patients with NSCLC who receive chemotherapy or chemotherapy combined with immunotherapy followed by radiotherapy; thus, there is no clear optimal treatment plan. If these patients use the same CTV margin, it may increase their treatment toxicity and affect their quality of life, which is also contrary to the current personalized and precise treatment strategy.

Therefore, this study aimed to use postoperative anatomopathological slices to evaluate the expansion of NSCLC beyond the macroscopic visible tumor after NAC and NACI, so as to determine the optimal CTV margin for patients receiving chemotherapy or chemotherapy combined with immunotherapy.

ME is considered as the “gold standard” in radiotherapy for accurately defining the CTV (17). Previous studies have quantitatively evaluated the subclinical invasion of primary lesions into surrounding tissues in untreated NSCLC to plan external beam radiotherapy (4, 18–20). However, to date, there is extremely little data on the ME of NSCLC after neoadjuvant therapy, which has resulted in a lack of unified and standardized guidelines to recommend the delineation of CTV for patients with NSLCL, who have received chemotherapy or chemotherapy combined with immunotherapy.

To the best of our knowledge, this is the first time in literature, that the macro pathology of NSCLC after chemotherapy and chemotherapy combined with immunotherapy was analyzed and ME distance beyond distinguishable GTV margins quantified. There is limited data available to directly determine whether CT tumor size fully represents gross pathological tumor size in NSCLC (18). To ensure the feasibility of our methodology, we first analyzed the relationship between tumor size measured using imaging and pathology. We confirmed that radiologic tumor size was closely related to macro pathological size without ME, similar to the conclusion of previous studies (4, 18, 19).

Several previous articles have published the ME values or the optimal CTV margins related to lung cancer radiotherapy. Giraud et al. (4) showed that to cover 95% of ME, the GTV of ADC and SCC required an outward expansion margin of 8 mm and 6 mm, respectively. Li et al. (20) determined that in order to ensure coverage of 95% of the ME when depicting the CTV, the required margins were 7 mm and 5 mm for ADC and SCC, respectively. ME measurements in these two studies were conducted using a light microscope. In our study, ME was accurately measured using a digital microscope, which may account for the discrepancy between the ME in the DS group and their research results. Schmitt et al. (18) after analyzing patients with NSCLC with localized stage T1N0 or T2aN0 who had undergone surgical treatment, concluded that to achieve 95% ME coverage, the required margins for SCC and ADC were 4.4 mm and 2.9 mm, respectively. Notably, in the above study, patients did not receive any other antitumor treatment before surgery; therefore, their ME range may not be suitable for inferring the CTV after receiving chemotherapy or chemotherapy combined with immunotherapy. Charlie et al. (19) included patients with NSCLC receiving NAC in their study and found that 95% of the ADC receiving and not receiving NAC had ME less than 5.3 mm (6.5 mm in the subgroup not receiving NAC), and 95% of the SCC receiving and not receiving NAC had ME less than 3.5 mm (3.7 mm in the subgroup not receiving NAC). Unfortunately, the number of patients receiving neoadjuvant chemotherapy included in their study was very small, only 28 cases, and there was no separate analysis of ME for this group of patients. Hu et al. (21) prospectively enrolled eight patients with small cell lung cancer (SCLC) and found that in patients who did or did not undergo NAC, a margin of 1.4 mm and 10.2 mm could cover 95% of ME, respectively. However, their research was restricted to limited-stage SCLC; thus, their conclusions were not applicable to NSCLC.

Regarding ME of tumors, the present study determined that the distribution of ME varied among different individuals with NSCLC, which correlated with the tumor pathological type, tumor stage, and preoperative treatment methods. It is noteworthy that the pathological type and preoperative treatment methods were the two major factors that significantly affected the ME distance. Hence, we defined the margin of CTV in NSCLC based on treatment methods and pathological types. To cover 95% of subclinical lesions, 6.80 mm and 6.00 mm CTV margins are recommended for direct surgical ADC and SCC, 5.20 mm and 4.20 mm CTV margins are recommended for ADC and SCC receiving chemotherapy before radiotherapy, and 4.60 mm and 1.20 mm CTV margins are recommended for ADC and SCC receiving chemotherapy combined with immunotherapy before radiotherapy.

Our research results indicate that there is a substantial difference in the microscopic infiltration range between ADC and SCC, with ADC demonstrating a larger microscopic infiltration range than that of SCC, consistent with the findings reported by Giraud et al. (4). Kara et al. (22) revealed that the infiltration of tumor cells in the central bronchus is more common in SCC than in ADC. However, the infiltration range of SCC is smaller than that of ADC, which may be correlated with the more invasive biological characteristics of ADC. Angela et al. (23) highlighted that spread through air spaces, mainly observed in lung ADC, has been identified as a novel mechanism of invasion, and which may be one of the reasons for the difference in ME between ADC and SCC. We also found that the mean tumor size of patients with SCC who did not undergo preoperative treatment was larger than that of patients with ADC using both imaging and pathology, while patients with SCC who underwent preoperative neoadjuvant treatment had a smaller average tumor size than that of patients with ADC; moreover, after neoadjuvant treatment, the PCR rate of SCC was higher than that of patients with ADC. SCC is more sensitive to current neoadjuvant therapy strategies than is ADC, and its potential benefits may be better (24, 25). Furthermore, this study demonstrated that tumor staging is related to ME size, which is consistent with the differences in ME research results of Schmitt et al. and Giraud et al. on NSCLC in stages I and I–IV, respectively (4, 18). This study also concluded that the ME size is related to the preoperative treatment received by the patients. The ME distance of the NAC group is smaller than that of the DS group, which may be due to the fact that chemotherapy drugs kill tumor cells by inhibiting key molecules and necessary cellular structures during cell replication (26). The ME distance of the NACI group is smaller than that of the NAC group. This is because programmed death receptor 1 (PD-1) and programmed death ligand 1 (PD-L1) inhibitors bind to PD-1 and PD-L1, respectively, preventing their interaction, restoring the recognition and killing effect of immune cells, avoiding immune escape of tumor cells, and thereby exerting anti-tumor effects (27). On the other hand, chemotherapy drugs can enhance anti-tumor immune responses through various mechanisms and to some extent inhibit the formation of immunosuppressive microenvironments (28–31). Previous studies have reported that NAC has better survival benefits than surgery alone (5–9), whereas subsequent studies have confirmed that NACI has better efficacy than NAC (10–12), which is consistent with the conclusions drawn from our study.

This study has some limitations. First, after the surgical specimen is removed from the human body, the tissue size may change, leading to inaccurate ME measurements. The surgical specimens were not fixed by the perfusion method because there is currently no experimental technique in literature that can make lung tissue specimens that leave the human body, completely consistent in morphology and structure similar to that of those inside the human body. Second, our analysis did not control for confounding factors which may exist, such as the types of drugs used for neoadjuvant therapy, dosage, treatment cycle, and scope of specimen collection; thus, further exploration is needed in the future.

This research aimed to personalize the delineation of the CTV in precise radiotherapy, thereby increasing the radiation dose to tumors while reducing the radiation damage to normal tissues, which is the primary goal of precision radiotherapy. We measured ME in NSCLC using three different treatment methods to define the CTV for radiotherapy. We observed that the mean ME of samples after chemotherapy or chemotherapy combined with immunotherapy (especially the latter) was much lower than the commonly adopted CTV range and differed considerably between the two histological types we studied. Therefore, the CTV currently used in clinical practice may increase therapeutic toxicity in patients. In our study, the mean ME values of ADC and SCC in the DS group were 1.98 mm and 1.26 mm, respectively. The mean ME values of ADC and SCC in the NAC group were 1.58 mm and 0.84 mm, respectively. The mean ME values of ADC and SCC in the NACI group were 0.87 mm and 0.14 mm, respectively. In the more practical context of radiotherapy, these findings can be represented in terms of the probability in which the ME corresponds to a given margin, hinging on the acceptable error risk. In the case of accepting a 5% error risk, for example, it would be essential to apply a margin of 6.80 mm and 6.00 mm to the ADC and SCC of DS, respectively; a margin of 5.20 mm and 4.20 mm to the ADC and SCC receiving chemotherapy, respectively; and a margin of 4.60 mm and 1.20 mm to the ADC and SCC receiving chemotherapy combined with immunotherapy, respectively.