For rectal cancers, determination of optimal irradiation targets based on their location and invasion pattern is a critical challenge. Despite the theoretical risk that tumor cells in LALRC with ACI can spread to the ILN region, there is no uniform agreement on whether the nodal region should be included in the CTV for this patient subgroup. More clinical evidence is required to optimize the CTV delineation for these patients in order to reduce irradiation of normal tissues.
The low ILN failure rate (4.94%) in our study, which mirrored other retrospective studies' findings, showed that most patients with LALRC with ACI would not benefit from elective inguinal irradiation during neoadjuvant or adjuvant (chemo)RT. Some experts, however, still recommends elective ILN irradiation based on acceptable morbidities. In our study, the acute toxicity associated with inguinal irradiation cannot be neglected. There were more acute grade 3 perineal dermatitis among patients who received elective inguinal irradiation (44.4% vs. 14.8%), though none required treatment break. Meanwhile, the reported chronic complications of elective inguinal irradiation appeared minimal in our study. Only 1 out of 9 patients (11.1%) who had elective inguinal irradiation developed a protracted gap wound after perineal recurrence.
Measures were developed to identify patients who were at a higher risk of developing inguinal nodal metastasis. Firstly, Maxiaowei Song et al. created a normogram to predict the probability of ILN failures according to tumor location, histological grade, and presence of perineural invasion8. It can be used as a guide to choose patients for elective inguinal irradiation who have a high chance of ILN failure, but the presence of perineural invasion can only be known after operation. Secondly, PET-CT was suggested to find out patient with inguinal uptake for inguinal nodal region irradiation. FDG-uptake of inguinal nodes at PET/CT may be present in up to 17% of patients with distal rectal cancer, particularly with ultra-low tumours [17]. Nevertheless, false positivity rate may be high, as nearly half of these nodes no longer demonstrated uptake after CRT despite that the inguinal region was not included in the radiation field [17]. Moreover, none of these patients in that study developed inguinal recurrence after 22 months of follow up [17]. A review of sentinel nodes in anal cancer found out that 44% of all node metastases located in lymph nodes measured less than 5 mm in diameter [18]. The spatial resolution of PET/CT is limited to several millimeters. Therefore, PET/CT may not have good enough sensitivity and specificity to select out patients for elective inguinal irradiation. Thirdly, the sentinel node technique was also investigated in rectal cancer with ACI. One small prospective study of 15 patients found that there was no inguinal recurrence for patient with sentinel nodes identified as negative for metastatic adenocarcinoma [19]. However, a systemic review commented that the sentinel-lymph-node procedure showed only a fair sensitivity rate of 82% (95% CI 60% to 93%), regardless of T stage, localisation or pathological technique [20]. Due to the relatively low sensitivity, technically demanding procedures, risk of surgical morbidity and doubtful impact to subsequent clinical management, this is not a standard practice for LALRC with ACI at the moment.
Only one patient (25%) developed isolated ILN metastases among all the 4 patients with inguinal recurrence. Salvage treatment for isolated ILN recurrence can provide long-term ILN control in our study. As a result, prophylactic treatment of the inguinal region may not be necessary. The other three patients (3 out of 4, 75% who experienced inguinal recurrence) had synchronous locoregional and/or distant recurrences. One may question whether early detection and treatment of occult inguinal nodal metastases can help prevent subsequent distant metastases. D. C. Damin et al observed that despite inguinal dissection, 75% of sentinel inguinal lymph node positive cases developed hepatic or pulmonary metastases within 6 months of the surgery [19]. Thus, localized treatment of the inguinal region may not influence the final clinical outcome, which is mainly determined by the emergence of metastases in distant organs [19]. In this context, a SLN metastasis could represent a potential marker for systemic dissemination of the disease [19].
From our results, patients who had positive pathological lymph node(s) following neoadjuvant therapy and/or a positive resection margin had an inferior rate of 3-year LRFS, DMRFS, and OS. It could imply that more aggressive neoadjuvant treatment is needed to shrink the tumor before surgery, such as the addition of an induction or consolidation chemotherapy regimen. Several recently published large scaled randomized controlled trials consistently showed that total neoadjuvant treatment can improve DFS, pathological complete remission rate, and reduce the risk of disease-related treatment failure in patients with high risk rectal cancer [21-24]. Among them, the phase 3 STELLAR trial was the first trial to demonstrate OS benefit, which found that short-course RT followed by perioperative chemotherapy resulted in better 3-year OS rates than CRT followed by postoperative chemotherapy, with 86.5% vs 75.1% (HR, 0.67; 95% CI, 0.46-0.97; P =.033) [24]. Furthermore, in our study, as compared to radiation alone, concomitant chemotherapy was linked with a superior LRFS. This was consistent with a Cochrane comprehensive review, which found that preoperative CRT improved local control (OR 0.56, 95% CI 0.42-0.75, P<0.0001) in resectable stage II and III rectal cancer but did not increase OS (OR 1.01, 95%CI 0.85-1.20, P=0.88) [25].
Song M et al. also investigated the impact of excluding irradiation of ILNs during neoadjuvant (chemo)RT in LALRC with ACI [8]. Their 3-year inguinal ILN failure rate was 3.7%. Our 3-year DMRFS was comparable to their study (77.0% vs. 77.7%), but our 3-year LRFS (81.1% vs. 94.6%) and OS (86.8% vs. 91.9%) outcomes appeared slightly inferior. Reasons for our relatively inferior locoregional and overall survival may be multifactorial. Our research population had an older median age (67 years old vs. 57 years old). Our study also covered a small number of patients with worse performance status (ECOG2) (6.7%) whereas their study only included ECOG 0-1 individuals. Almost all of our patients (93.3%) received bolus 5-FU as concurrent chemotherapy, with the exception of one patient who received oral capecitabine, compared to 78.1% of capecitabine patients in their study. Patients who received a protracted infusion of 5-FU had a significantly longer time to relapse and improved survival when compared to bolus 5-FU [26]. Two randomized controlled trials have shown that rectal cancer patients who received neoadjuvant or adjuvant capecitabine CRT had non-inferior disease free and overall survival when compared to continuous 5-FU [15,16]. Therefore, concurrent chemotherapy with oral capecitabine should produce better outcomes compared with bolus 5-FU. In addition, induction (7.0%) and consolidation chemotherapy (30.8%) were used in their research, which might further boost treatment outcomes.
Our study had several limitations. To begin with, it was a retrospective study based on data from a single center, and this may add selection and information biases. Secondly, our small sample size reduced the power of the study. Thirdly, as there was no uniform follow-up imaging in our study population, the recurrence-free interval and OS may have been overstated.