According to the National Comprehensive Cancer Network (NCCN) guidelines, neoadjuvant chemoradiotherapy followed by surgery (NACRT-S) is recommended for patients with resectable superior sulcus tumors, and is an alternative option for patients with resectable stage IIIA non-small cell lung cancer (NSCLC) (1). However, neoadjuvant therapy is a major risk factor for postoperative broncho-pleural fistula (BPF) and respiratory failure (2). From the perspective of radiotherapy (RT), several studies have reported dosimetric risk factors using dose–volume histogram (DVH) parameters related to pulmonary complications after NACRT-S for NSCLC (3-6). In definitive chemoradiotherapy for locally advanced NSCLC, intensity-modulated radiotherapy (IMRT) has been more useful than three-dimensional conformal radiotherapy (3D-CRT) in reducing lung toxicities (7,8). The usefulness of IMRT in the setting of NACRT-S remains unknown. Therefore, in this planning study, we compared 3D-CRT with IMRT to avoid the dosimetric risk factors of NACRT-S for NSCLC.

Patients. This retrospective study was approved by the institutional ethics committee (approval number: 2019-053). The eligibility criteria for this study were as follows: 1) patients who underwent NACRT-S for NSCLC between 2016 and 2019 at our institution; 2) patients who underwent 3D-CRT with a dose of 50 Gy in 25 fractions; and 3) patients with dosimetric risk factors in the clinical treatment using NACRT-S.

Radiation treatment planning. Four-dimensional computed-tomographic images of free breathing were obtained for radiation treatment planning. Primary tumors and clinically positive lymph nodes (LNs) were defined as the gross tumor volumes (GTVs). The clinical target volume (CTV) was defined as the GTV with a 5 mm margin and the nodal station to which the clinically positive LNs belonged. Elective nodal irradiation was not performed to the non-metastatic stations. The planning target volume (PTV) was defined as the CTV with a 5 mm margin.

Regarding the DVH parameters, we calculated the percentage of the lung volume that received more than x Gy (V_x) and the mean lung dose (MLD) using: 1) the total lung volume minus the GTV (DVH_g: Figure 1A); 2) the lung volume remaining after surgery (DVH_r: Figure 1B); and 3) the contralateral lung volume (DVH_c: Figure 1C). Previous reported dosimetric risk factors in NACRT-S (3-6) were as follows: 1) for DVH_g, V_20g≥21%, V_20g≥23%, V_35g≥19%, V_40g≥16%, MLD_g≥10 Gy, and MLD_g≥10.8 Gy; 2) for DVH_r, V_20r≥10%, V_20r≥12%, and MLD_r≥5.6 Gy; and 3) for DVH_c, V_10c≥20% and V_20c≥7%.

In this study, we generated radiation treatment plans using 3D-CRT and IMRT to avoid the aforementioned dosimetric risk factors for the lungs and achieve common dose constraints for the spinal cord based on the NCCN guidelines (1). We were careful not to broaden the low doses to the lungs during beam arrangement. The prescribed doses that were normalized to 50% of the volume of the PTV were 50 Gy in 25 fractions.

Statistics. We analyzed the dosimetric differences between 3D-CRT and IMRT using the Wilcoxon rank-sum test and Fisher’s exact test. A p-value of <0.05 was considered to indicate statistical significance. Statistical analyses were performed using JMP Pro ver. 15 (SAS Institute, Cary, NC, USA).

Eleven patients met the eligibility criteria, and their tumor characteristics are listed in Table I. Dosimetric differences between 3D-CRT and IMRT are listed in Table II. V_35g and V_40g were significantly lower with IMRT than with 3D-CRT (p=0.001 each). As for each dosimetric risk factor (Table III), 9% and 55% of the patients were able to avoid MLD_g≥10 Gy with 3D-CRT and IMRT, respectively (p=0.032). Overall, 0% and 55% of the patients were able to avoid all dosimetric risk factors with 3D-CRT and IMRT, respectively (p=0.006).

Additional analyses were performed for IMRT with or without avoiding all dosimetric risk factors (Table IV). Tumor location and the length of PTV significantly affected the avoidance of all dosimetric risk factors (p=0.015 and 0.022, respectively). As for the relationship between PTV length and tumor location, the PTV length of the middle or lower lobe tumors was significantly longer than that of the upper lobe tumors: median, 14.2 cm and 10.0 cm, respectively (p=0.013).

NACRT-S is a well-known risk factor of pulmonary complications during NSCLC treatment (2,11). To reduce the toxicity, improvements were explored from the perspective of RT. Regarding the dosimetric risk factors of NACRT-S for NSCLC, V_20r≥12%, V_35g≥19%, and V_40g≥16% were first proposed as significant factors affecting the incidence of radiation pneumonitis (RP) and BPF or pulmonary fistulas (3). Second, V_20r≥10% and MLD_r≥5.6 Gy have been reported to be significant predictors of RP (4). Third, V_10c≥20% and V_20c≥7% were significant factors affecting the incidence of pulmonary toxicity (5). Finally, V_20g≥21% and MLD_g≥10 Gy have been reported to be significant predictors of RP (6).

In this planning study, we compared 3D-CRT with IMRT to avoid the dosimetric risk factors associated with NACRT-S for NSCLC. IMRT significantly reduced V_35g and V_40g compared to 3D-CRT. Among the dosimetric risk factors, MLD_g≥10 Gy was significantly avoided using IMRT. Overall, 0% and 55% of the patients were able to avoid all dosimetric risk factors with 3D-CRT and IMRT, respectively. In the treatment of definitive chemoradiotherapy for locally advanced NSCLC, IMRT has been useful in reducing the radiation dose to the lungs compared to 3D-CRT (7,8). We confirmed that IMRT is more useful than 3D-CRT in reducing the irradiated dose to the lungs in the NACRT-S setting, as with the definitive setting.

However, even with IMRT, all of the four patients with middle or lower lobe tumors had at least one dosimetric risk factor. Tumor location and PTV length significantly affected the avoidance of all dosimetric risk factors. The PTV of the middle and lower lobe tumors was significantly longer than that of the upper lobe tumors. One possible reason for this is the respiratory motion of the tumor. Respiration-induced tumor motion in the superior-inferior direction was greater in the lower lobe and lower pulmonary zone tumors compared with apical tumors (9). Respiratory motion management is recommended to spare the normal tissue when patients can tolerate the procedure (10). To further improve the avoidance of dosimetric risk factors for NACRT-S, we should apply additional respiratory motion management, such as breath-holding, for patients with middle or lower lobe tumors.

Our study has some limitations, such as the small number of participants and the retrospective single-institutional design.

In conclusion, our findings suggest that IMRT is more useful than 3D-CRT for avoiding dosimetric risk factors in NACRT-S for NSCLC. For further improvements in avoiding these factors, respiratory motion managements to reduce the length of the PTV may be required for patients with middle or lower lobe tumors.

The Authors have no conflicts of interest regarding this study.

This study was coordinated by ST and TS. The data were collected by ST, MA, TK, and TN. The collected data were analyzed by ST. This article was drafted by ST. Data interpretation and article revision were performed by all authors: ST, MA, TK, TN, and TS. All the Authors have approved the submitted manuscript.

This work was partly supported by the Young Scientists Fund of Kagawa University Research Promotion Program 2019 (KURPP) and JSPS KAKENHI Grant Number JP20K16790.