Total Neoadjuvant Chemotherapy Generates Comparable Efficacy to Chemoradiotherapy in Locally Advanced Rectal Cancer

Short-term radiotherapy followed by total neoadjuvant chemotherapy (TNT) was found to be noninferior to long-term chemoradiotherapy (CRT) in patients with locally advanced rectal cancer, according to findings from the phase 3 STELLAR study (NCT02533271) published in the Journal of Clinical Oncology.

At a median follow-up of 35.0 months, the 3-year disease-free survival (DFS) rate was 64.5% (95% CI, 58.3%-70.7%) and 62.3% (95% CI, 56.1%-68.5%) in TNT and CRT groups, respectively (HR, 0.883; one-sided 95% CI, not applicable-1.11; P < .001 for noninferiority). Although there was no significant difference in metastasis-free survival or locoregional recurrence between the two groups, the TNT group exhibited better 3-year overall survival (OS) rates than the CRT group, at 86.5% (95% CI, 82.1%-90.8%) vs 75.1% (95% CI, 69.4%-80.8%), respectively (HR, 0.67; 95% CI, 0.46-0.97; P = .033).

These results suggest that the combination of short-term radiotherapy, preoperative chemotherapy, and surgery is a viable alternative to CRT in terms of efficacy and acceptable toxicity for locally advanced rectal cancer.

The current first-line treatment for locally advanced rectal cancer is long-course CRT followed by total mesorectal excision (TME). Only approximately 50% of patients who undergo subsequent adjuvant chemotherapy finish this therapy. Neoadjuvant short-course radiotherapy (5 Gy in 5 fractions) is another treatment option for resectable locally advanced rectal cancer that has been shown to have comparable outcomes to long-course CRT in terms of OS, DFS, local control, and late toxicity.

The STELLAR study aimed to investigate whether neoadjuvant chemotherapy plus short-term radiotherapy in place of CRT, combined with surgery and adjuvant chemotherapy, could improve patient compliance and completion rates during chemotherapy and enhance treatment efficacy in locally advanced rectal cancer.

The study enrolled patients ages 18 to 70 years with ECOG scores of 0 or 1. Eligible patients had distal or middle-third, clinical primary tumor (cT) stage III to IV, and/or regional lymph node–positive rectal adenocarcinoma.

Patients with a history of previous anticancer treatments, recurrent disease, second primary malignancy, or medical contraindication to the planned treatment or magnetic resonance imaging (MRI) were excluded.

Eligible patients underwent preliminary pelvic MRI to determine mesorectal fascia (MRF) involvement, defined by the tumor distance of the MRF being less than 1 mm regardless of metastatic lymph nodes, a primary tumor, or MRI-extramural vascular invasion (EMVI). In addition, patients underwent digital rectal examination, colonoscopy with biopsy, an optional endorectal ultrasound, chest computed tomography (CT), liver MRI and CT, and biochemical examination with serum carcinoembryonic antigen. Patients were re-evaluated with these same examinations 5 to 6 weeks after neoadjuvant therapy.

Between August 30, 2015, and August 27, 2018, 591 patients were treated across 16 hospitals in 11 provinces of China.

A total of 298 patients in the TNT group were assigned to receive short-term radiotherapy (5 Gy in 5 fractions) followed by chemotherapy (4 cycles of CAPOX: oxaliplatin 130 mg/m2 once daily, on day 1 and capecitabine [Xeloda] 1000 mg/m2 twice daily, from day 1 to 14) 1 to 2 weeks after radiotherapy completion, followed by surgery and 2 cycles of adjuvant CAPOX. The other 293 patients in the CRT group were assigned to receive long-term chemoradiotherapy (50 Gy in 25 fractions) over 5 weeks concurrently with capecitabine (825 mg/m2, twice daily), followed by surgery and 6 cycles of CAPOX.

DFS served as the primary end point of this study, defined as the time from the date of randomization to the first occurrence of distant metastases (DM), second primary tumor, locoregional failure, or death from any cause. Secondary end points included OS, metastasis-free survival (MFS), locoregional recurrence, and surgical complications.

All patients in the TNT group completed radiotherapy without dose reduction. In the CRT group, 1.4% of patients experienced radiation dose reduction, and 2.4% of patients experienced interruption of radiotherapy. In the CRT group, 7.8% of patients experienced chemotherapy dose reduction.

Across both groups, the neoadjuvant treatment completion rates (including reduced radiotherapy or chemotherapy doses, or delayed completion) were 82.6% and 95.2% (P < .001) in the TNT and CRT groups, respectively. The respective full-dose neoadjuvant treatment completion rates were 74.8% and 93.2% (P < .001) in the TNT and CRT groups.

Regarding safety during neoadjuvant treatment, 26.5% of the TNT group experienced acute grade 3 to 5 toxicities vs 12.6% of the CRT group (P < .001). The most prevalent grade 3 to 4 acute toxicity was hematologic, which affected 15.8% of the TNT group vs 2.0% of the CRT group (P < .001).

Upon re-evaluation with MRI, digital rectal examination, and colonoscopy after neoadjuvant therapy, 11.1% (n = 33) of the TNT group and 4.4% (n = 13) of the CRT group achieved clinical complete response (cCR), regardless of watch-and-wait or surgery. A total of 9.4% (n = 28) of the TNT group and 3.4% (n = 10) of the CRT group achieved cCR after neoadjuvant treatment and did not undergo additional surgery.

Of the patients who did not achieve cCR, 17 in the TNT group and 36 in the CRT group refused surgical procedures for personal reasons. In total, 235 patients in the TNT group and 230 patients in the CRT group received primary tumor resection.

The 2 groups experienced similar percentages of grade 3 or greater surgical complications: 14.0% of patients in the TNT group and 15.7% of patients in the CRT group (P = .625). The predominant postoperative complication was delayed wound healing. There were no deaths within 30 days post operation.

Of the 2 groups, 465 patients completed surgery: 235 in the TNT group and 230 in the CRT group. In the TNT group, 91.5% of patients achieved R0 resection compared with 87.8% of patients in the CRT group (P = .189).

In the populations that received surgery, 23.0% (n = 54) of patients in the TNT group did not receive adjuvant chemotherapy compared with 26.1% (n = 60) in the CRT group. Of the 351 patients who received adjuvant chemotherapy, 60.0% (n = 108) of the TNT group and 48.3% (n = 82) of the CRT group completed planned cycles (P = .009). In terms of safety, 3.3% of the TNT group who received adjuvant chemotherapy experienced grade 3 to 5 toxicities compared with 11.8% of the CRT group (P = .003).

In total, 21.8% of the TNT group experienced either sustained cCR or pathological complete response (pCR) compared with 12.3% of the CRT group (P = .002).

At the median follow-up of 35.0 (range, 8.3-63.9) months, throughout the intention-to-treat (ITT) population (n = 599), 202 patients (99 in the TNT group and 103 in the CRT group) experienced locoregional recurrence, metastasis, or death as a result of any cause.

Findings from the subgroup analysis showed no significant difference in treatment effects on progression-free survival (PFS) and OS regardless of clinicopathologic prognostic factors.

The authors concluded that the combined findings from the STELLAR study provide evidence that treatment with TNT is at least as favorable as CRT in terms of survival, locoregional control, compliance, and tolerability and has potential as an alternative to CRT and adjuvant chemotherapy in patients with middle and low locally advanced rectal cancer.

Going forward, longer follow-up with TNT and CRT at 5 to 10 years is needed to determine the long-term effects of TNT on OS and other clinical outcomes.

Reference

Jin J, Tang Y, Hu C, et al. Multicenter, randomized, phase III trial of short-term radiotherapy plus chemotherapy versus long-term chemoradiotherapy in locally advanced rectal cancer (stellar). J Clin Oncol. Published online March 9, 2022. doi:10.1200/JCO.21.01667