Concurrent paclitaxel-based chemo-radiotherapy for post-surgical microscopic residual tumor at the bronchial margin (R₁ resection) in non-small-cell lung cancer

To our knowledge, this is the first report of concurrent CRT for post-surgical microscopic residual tumor at the bronchial margin in patients with NSCLC. Paclitaxel-based chemotherapy and concurrent radiotherapy achieved a median OS of 32.0 months among those NSCLC patients after R1 resection, with tolerable treatment-related toxicity. Although this was a retrospective evaluation with a small sample, our data suggest that selected NSCLC patients after R1 resection may benefit from aggressive and curative-intent concurrent CRT.

Reports from Liewald et al. and Snijder et al. show that for Stage I and II NSCLC patients with positive margins, repeat resection could improve OS [3,4] and is recommended by NCCN [1] for Stage II patients. In the present study, 11 patients (23.9%) with Stage II NSCLC did not undergo repeat resection, but received CRT. As recorded in our database, four patients refused reoperation because of limited cardiopulmonary function, and the others did not want a second operation. Among these patients, the median survival time was 35.2 months, which was similar to the reported data of repeat resection in Stage II cases by Liewald et al. (median OS: 38 months) [3].

The NCCN panel recommends the concurrent CRT for R₂ resection or mediastinal recurrence, and the sequential CRT for R₁ resection [1]. However, there is no direct evidence of any disadvantage of concurrent settings in NSCLC patients with R₁ resection. Concurrent chemoradiation improves the clinical outcomes of Stage IIIA or IIIB disease [13-15]. In the present study, salvage CRT was well-tolerated and toxicity was as expected from thoracic CRT. The median OS was 32.0 months, which is comparable to those data collected from patients with bronchial stump recurrence [16]. Recently, Bar et al. reported the outcomes of CRT for loco-regional recurrence of NSCLC after surgery, and the median survival after recurrence was 26.9 months [17]. Thus, for patients with good ECOG performance status after R₁ resection, concurrent CRT is still a treatment of choice.

It should be mentioned that three patients (6.5%) developed acute grade 3 radiation pneumonitis after treatment. Two patients had received right lower lobectomy and one left lower lobectomy. The delivered dose was 60, 62 and 60 Gy respectively, and the chemotherapy regimen was paclitaxel and cisplatin. They were diagnosed with Grade 3 radiation pneumonitis between 2 and 4 weeks after radiotherapy and finally recovered after steroid therapy. The incidence of RP is somewhat lower than the concurrent CRT for locally advanced NSCLC. Among several parameters based on dose-volume histograms, V_dose and mean lung dose (MLD) are important predictive factors of acute radiation pneumonitis [18,19]. In definitive chemo-radiotherapy for NSCLC, the NCCN panel suggests that a V₂₀ value of 30-35% and MLD <20 Gy are thresholds for symptomatic radiation pneumonitis [1]. However, there is little information of such a threshold in the post-lobectomy situation. Uno et al. reported that in a 21-patient population, three patients developed grade ≥2 radiation pneumonitis after concurrent CRT [20]. They found that the V₂₀ < 20%/MLD <10 Gy might be predictive factors for grade ≥2 radiation pneumonitis in post-lobectomy patients receiving definitive radiotherapy. In our practice, the lung constraints were set as V₂₀ < 22% and MLD <12 Gy. All these constraints need more studies for validation.

In this study, the drug doses were modified in 19 (41.3%) patients during treatment. In addition, the granulocyte colony stimulating factor (G-CSF) is routinely applied for secondary prophylaxis in our practice when patients have grade 1 or 2 neutropenia in the preceding cycle. In some situations, we use G-CSF prophylactically among patients after chemotherapy to avoid any break in radiotherapy. So, the rate of grade 4 neutropenia (1 patient, 2.2%) was lower than expected. No grade ≥3 acute esophagitis was observed in this study. To avoid any break in radiotherapy as a consequence of grade 3 acute esophagitis, we usually prescribe a liquid combination (500 ml 0.9% physiological saline injection, 10 ml 1% lidocaine injection, and 10 mg dexamethasone mixed with each other, 15 ml P.O three times per day) among the patients with acute grade 2 esophagitis.

Another issue that should be discussed here is the target delineation. As all tumors are microscopic at the bronchial margin, it was difficult to define the GTV. Griess et al. [21] and Cotton [22] have reported that even after resection, in which there is a macroscopic tumor-free margin >2 cm, the incidence of R₁ resection was still around 6% among these resections. From the report by Olszyna-Serementa et al., 80 patients with R₁ resection have been analyzed [23]. They concluded that the PORT results in a relatively better survival in these patients. They also suggested that the elective nodal irradiation was useful for local control in pN_0–1 patients. At present, the precise definition of GTV and CTV in an R₁-resection situation had not been concluded and needs more clinical investigation.

By univariate survival analysis, we found that pathological type (non-squamous cell lung cancer) was significantly longer survival time, compared with patients with squamous-cell lung cancer (45.1 vs 26.4 months, p = 0.013). This result differs from the studies of Ghiribelli et al. [5] and Liewald et al. [3]. This may in part be explained by imbalances in the selection of subsequent treatment regimens with respect to histological subtypes. Currently, it is well known that several new anti-tumor drugs (including pemetrexed and TKIs) could significantly prolong the survival time among patients with metastatic non-squamous cell lung cancer, since a series of landmark trials has been published [24-27]. It might be the reason that these patients survive longer than the others in the present study.

Limitations of the present study should be mentioned. First, the retrospective nature of the study and the small number of the patients must be paid attention when interpreting the results. Second, the patients analyzed in this study had good ECOG performance status (0 or 1). Some patients could not tolerate the adjuvant CRT or chemotherapy alone, if their performance status was 2 or 3.