NAR score predicts prognosis in Chinese patients with locally advanced rectal cancer after neoadjuvant CRT

A retrospective study in Hong Kong has shown that neoadjuvant rectal (NAR) score is an independent prognostic marker in Chinese patients with locally advanced rectal cancer (LARC) treated with neoadjuvant chemoradiotherapy (CRT), indicating its potential applicability in clinical trials and routine practice.

NAR score is a composite endpoint calculated based on the changes in clinical and pathological tumour stages, plus nodal status after neoadjuvant CRT, resulting in 24 possible discrete scores ranging from 0 to 100. Lower NAR scores represent better prognosis. [Curr Colorectal Cancer Rep 2015;11:275-280; Lancet Oncol 2021;22:e314-e326]

“The NAR score has been shown to better predict overall survival [OS] than pathological complete response [pCR, a binary endpoint defined as absence of visible residual tumour after neoadjuvant therapy] in clinical trials in Western populations,” the authors noted. “This study aims to validate the prognostic significance of NAR score and factors associated with lower NAR scores in Hong Kong Chinese patients with LARC.” [Hong Kong Med J 2022;28:230-238]

The total study population comprised 193 Chinese LARC patients (mean age, 62 years; male, 74.6 percent) who achieved optimal responses to neoadjuvant CRT and total mesorectal excision at the Prince of Wales Hospital between August 2006 and October 2018. All treatment decisions were made by the local multidisciplinary Lower Gastrointestinal Tumour Board.

With a median follow-up of 47.7 months, NAR discrete scores were associated with survival outcomes, including OS (hazard ratio [HR], 1.042; 95 percent confidence interval [CI], 1.021–1.064; p<0.0001), disease-free survival (DFS; HR, 1.042; 95 percent CI, 1.022–1.062; p<0.0001), locoregional recurrence-free survival (LRFS; HR, 1.070; 95 percent CI, 1.039–1.102; p<0.0001) and distant recurrence-free survival (DRFS; HR, 1.034; 95 percent CI, 1.012–1.056; p=0.002).

When NAR scores were categorized as low (<8; n=50), intermediate (8–16; n=99) or high (>16; n=44), associated 5-year OS rates were 97.8 percent, 75.2 percent and 56.0 percent, respectively, demonstrating improved OS rate inlower vs higher NAR score groups (p=0.004).

Similarly, the low NAR score group had significantly improved rates of DFS, LRFS and DRFS vs intermediate and high score groups (5-year DFS rate, 93.3 percent vs 73.7 percent vs 57.1 percent; p<0.0001) (5-year LRFS rate, 100 percent vs 92.2 percent vs 77.6 percent; p=0.002) (5-year DRFS rate, 93.3 percent vs 73.6 percent vs 67.4 percent; p=0.013).

In patients with intermediate NAR scores, late recurrence, particularly distant recurrence was observed 72 months after diagnosis. “Longer follow-up duration for >72 months may be required [in patients with intermediate NAR scores],” the authors suggested.

Among prognostic factors analyzed, only pCR was associated with lower NAR scores (≤8 vs >8; p<0.0001). On the other hand, MRI tumour down-staging was the only prognostic factor associated with pCR (p<0.0001).