Linking systemic inflammation nutrition status to the tumor immune microenvironment: integrated biomarkers of neoadjuvant response in locally advanced rectal cancer

Abstract

Immune checkpoint inhibitors (ICIs) have transformed the treatment landscape of colorectal cancer (CRC), yet response remains heterogeneous. The inflammation-nutrition prognostic score (INPS), defined as the neutrophil-to-lymphocyte ratio (NLR) divided by the prognostic nutritional index (PNI), integrates systemic inflammatory burden and nutritional-immunological reserve and may provide complementary information in locally advanced rectal cancer (LARC). We retrospectively analyzed 312 patients with LARC who received standard neoadjuvant chemoradiotherapy (nCRT), with or without concurrent PD-1/PD-L1 blockade, followed by total mesorectal excision (TME). Pretreatment INPS was calculated as NLR/PNI. The primary endpoint was objective response rate (ORR); secondary endpoints included pathological complete response (pCR) and progression-free survival (PFS). ROC, logistic regression, interaction, sensitivity, and Cox analyses were performed. Among 312 patients, 94 (30.1%) received ICI-containing regimens. Low INPS (<0.058) was associated with higher ORR (61.3% vs. 29.4%, P < 0.001) and pCR (38.7% vs. 11.8%, P < 0.001). INPS yielded an AUC of 0.765 (95% CI: 0.714–0.816) for ORR, compared with 0.712 for NLR, 0.668 for PNI, and 0.660 for PD-L1 CPS. In multivariable analysis, low INPS (OR = 3.31, 95% CI: 1.95–5.60, P < 0.001), MSI-H/dMMR status (OR = 4.67), PD-L1 CPS ≥ 10 (OR = 2.74), lower clinical T stage (OR = 3.15), and an ICI-containing regimen (OR = 2.55) were independently associated with response. A forced-entry disease-burden sensitivity model confirmed the association of low INPS (adjusted OR = 3.04, 95% CI: 1.74–5.30, P < 0.001). Low pretreatment INPS was independently associated with favorable response and PFS in LARC and provided incremental discrimination beyond MSI-H/dMMR and PD-L1 CPS. These findings support INPS as a candidate response-associated biomarker in LARC; prospective and external validation is warranted before routine clinical application.