This is the first study evaluating the incidence of pneumonitis, either by radiotherapy or by ICI, in patients with NSCLC who are treated with CCRT, followed by Durvalumab in two different total radiation dose cohorts.

The results showed that 39% of patients developed any type or grade of pneumonitis, with 35% in the 70 Gy group (70% RI‑P; 30% ICI-P) compared to 50% in the < 70 Gy group (80% RI‑P; 20% ICI-P). These finding are consistent with previous literature, such as the PACIFIC trial by Antonia et al., which reported a pneumonitis rate of 34%, with 11% attributed to immune-related pneumonitis [1]. Similar results were reported by Shintani et al. [15]. In our study, most patients developed grade 2 pneumonitis, indicating a moderate level of severity, while less patients experienced grade 3 pneumonitis. Those findings are again in line with other authors [1, 16]. Furthermore, in both groups, the median time to onset of any cases of pneumonitis was similar (3 vs. 4 months), which is also comparable with the literature [17].

A main finding is that the OS was significantly different between patients receiving 70 and < 70 Gy (p = 0.028). In the < 70 Gy group median OS was 31 months, whereas in the 70 Gy group the median OS was not yet reached. This suggests that patients in the 70 Gy group had a better OS outcome compared to those in the less than 70 Gy group. This might indicate a potential benefit associated with a higher radiation dose, despite other studies reporting the opposite [3,4,5,6]. Moreover, the < 70 Gy group demonstrated a higher incidence of tumor progression (60%) compared to the 70 Gy group (27.6%). However, these findings may be seen as a potential confounder because patients with high mean lung doses, which imposed a high dose constraint, were administered less than 70 Gy total tumor dose, despite of a larger tumor mass. In other words, this is due to the fact that the dose is meant for the tumor but not the lungs, and with very large tumors it may happen that the individual tumor dose is lower, otherwise it would exceed dose constraint for the lung. Furthermore, patients receiving < 70 Gy maybe were considered more co-morbid, or with lower performance score. However, ECOG (Eastern Cooperative Oncology Group) status, age, gender, NSCLC disease stage, tumor histology, and tumor location were well balanced and did not differ between the groups.

OS data were still immature at the time of analysis, but the 70 Gy group showed promising results. After 12 months, the OS rate was 93.1%, and it remained at almost 90% over the next two years. These findings are comparable to other studies such as the “Five-Year Survival Outcome from the PACIFIC trial” by Spigel et al. [18], or other reports with an 12 months OS between 86 and 92% [2, 19]. Other studies that used higher radiation dose reported worse overall survival compared to those who received the standard dose of 60 Gy [6]. However, these studies did not used Durvalumab for consolidation therapy and applied up to 74 Gy.

In this study, 31% of the patients completed the recommended one year of Durvalumab treatment without discontinuation. No patient was still on maintenance therapy at the time of analysis. These findings are consistent with current literature, which report of 29 or 42% completion rates of the ICI consolidation therapy [20, 21]. The most common reasons for discontinuation of Durvalamub were mainly pneumonitis, or disease progression. The Society for Immunotherapy of Cancer recommends considering drug re-challenge in patients with grade 2 pneumonitis that has completely resolved, as well as in selected patients with grade 3 pneumonitis [22]. In this analysis, a third of the patients who experienced pneumonitis and required treatment, re-entered ICI treatment successfully. No relapse of pneumonitis was observed, and only one patient had tumor progression after re-entering. Furthermore, 75% of patients were able to continue treatment after re-entering. This finding highlights the potential benefits of considering ICI re-challenge more frequently [10, 20, 23,24,25,26].

The planning target volume and MLD are crucial factors in determining the optimal total radiation dose for patients undergoing concurrent chemoradiotherapy for unresectable NSCLC [4]. In this study, patients with high MLD which led to high dose constraint, received less than 70 Gy total prescribed dose. This suggests that the consideration of lung dose constraints, particularly the MLD, influenced the decision to limit the total radiation dose to ensure the safety of patients. The adherence to dose constraints aims to minimize the risk of radiation-induced lung toxicity, including pneumonitis. Interestingly, we observed a significant difference in mean dose in the ipsilateral lung. This difference can be attributed to adapting treatment planning strategies and the effort to achieve adequate target coverage while minimizing lung exposure. Furthermore, the mean dose for the total lung did not exceed the safety criterion of MLD ≤ 20 Gy for both groups [27]. This indicates that, despite differences in mean dose between the groups, radiation doses to the entire lung were kept within acceptable limits to reduce the risk of radiation-induced lung toxicity. There were no significant differences in V20, V10 and V5 between the groups. However, total lung volume receiving more than 5 Gy (V5) was significantly higher in the 70 Gy group among patients with radiation-induced pneumonitis. Radiotherapy guidelines for NSCLC suggest V5 should be kept below 60% to minimize the risk of pneumonitis [28]. The V5 values in the 70 Gy group with RI‑P in this study exceeded the recommended threshold, which might has contributed to the appearance of a RI‑P. These results highlight the importance of considering not only the MLD but also specific dose-volume parameters in assessing the risk of pneumonitis.

An important limitation of this study is the very small sample size. Being conceptualized as a retrospective pilot study, the group size is especially low for patients receiving < 70 Gy total radiation dose. The study focuses on the appearance of pneumonitis, and the higher radiation dose was not inferior in this matter. Other evaluations, such as survival, must be interpreted very cautiously since the study was not powered accordingly. On the other side, this study provides interesting information and the data is new. Future prospective studies should be planned with sample size calculation and a possible randomization.

Since this was a retrospective design, randomization was not possible. Another issue in this context is that tumor size is considered a potential confounding factor, because radiation dose of the control group with < 70 Gy mainly depended on MLD, V20, V10, and V5 values. Randomization in future studies would reduce this possible limitation.

Another limitation is a potential misclassification bias in differentiating between ICI‑P and RI‑P, as the information regarding pneumonitis may be concealed within the radiation field.

There is an ongoing discourse on the probability of bias creation following the administration of antiangiogenic drugs due to the presence of antiangiogenetic food components. In this study, no data on dietary behavior was collected. Future studies should use a validated semi-quantitative food frequency questionnaire to gather information on these agents. [29].

Lastly, the inclusion criterion for the study was based on potentially completing one year of Durvalumab therapy. However, for survival analysis, the total duration of therapy since its initiation was collected, which varied for each patient. This difference in data collection may introduce bias due to the varying observation periods.