Comparison of Nutritional and Immunological Scoring Systems Predicting Prognosis before Anti-B Cell Maturation Antigen Chimeric Antigen Receptor-T Therapy in Multiple Myeloma, a Retrospective Study Based on Legend-2

Abstract Background Malnutrition is a widespread issue among people with hematological malignancies, and it is associated with worse outcomes. Emerging evidence indicates that poor nutritional status determined with nutritional indices such as controlling nutritional status score (CONUT), prognostic nutritional index (PNI), and naples prognostic score (NPS), was associated with poor prognosis of hematological malignancy. In terms of CAR-T therapy, as is a relatively novel treatment, there is a scarcity of research into its impact on nutritional status on patients. And there are no findings on using the controlling nutritional status (CONUT) score, prognostic nutritional index (PNI) and naples prognostic score (NPS) to evaluate the prognosis of patients receiving CAR-T therapy. Aims We conducted this retrospective study to quantitatively evaluate the nutritional status of patients before CAR-T treatment and its impact on the efficacy as well as prognosis, and compare prognostic values of the three indices in refractory or relapsed multiple myeloma (R/R MM) patients. Methods The clinical data of 56 patients with who underwent anti-B cell maturation antigen chimeric antigen receptor-T therapy at our hospital from 2016 to 217 were collected and analyzed retrospectively (NCT03090659). PNI, CONUT and NPS before CAR-T cell infusion were calculated by blood indicators, including absolute lymphocyte count (ALC), lymphocyte to monocyte ratio, neutrophil to lymphocyte ratio, serum albumin (ALB), and serum total cholesterol (TC). Receiver operating characteristic (ROC) curve was used to estimate the discriminative cutoff value of each nutritional index and patients were divided into high (PNI> 37.8) and low (PNI≤37.8) PNI groups and high (score＞4) and low (score≤4) CONUT groups. Patients were divided by NPS score into group 0, group 1, and group 2. The relationship between the three immune-nutritional scores and clinical characteristics was analyzed. Cox regression and Kaplan-Meier analysis were performed to analyze the difference in overall survival (OS) and progression free survival (PFS) between various immune-nutritional score groups. Fisher's exact test was performed to assess differences between three indices and efficacy. RESULTS Subtypes (P=0.0141) and ECOG performance status (P=0.0005) were significantly related to CONUT. ECOG performance status (P=0.0038) and time from initial MM diagnosis (P=0.0152) were closely related to PNI. Subtypes (P=0.0371), prior lines of therapy (P=0.0392) and previous autologous stem cell transplantation (P=0.0257) were all significantly associated with NPS. The Kaplan-Meier curve indicated that CONUT score (P=0.0388) and PNI (P=0.0264) were significantly associated with OS, but NPS results were not. The univariate analyses demonstrated that ECOG performance status (hazard ratio [HR], 1.710; 95% confidence interval [CI], 1.050-2.786; P=0.031), best response (HR, 1.921; 95% CI, 1.167-3.165; P=0.010) and nutritional status determined using CONUT (HR, 2.550; 95% CI, 1.195-5.441; P=0.016) were significantly associated with a poorer OS. Multivariate analysis showed that three nutritional indices did not persist as independent prognostic indicators. As for treatment efficacy, we found that overall response rates (ORRs) were significantly lower in patients from high CONUT group than in those from low CONUT group (P=0.0281). CONCLUSION This study demonstrated that CONUT and PNI are important for assessing the prognosis of R/R MM patients underwent CAR-T treatment. The results showed that poor immune-nutritional status was associated with a poor OS. Among the three immune-nutritional scoring systems, the CONUT scoring system can more accurately evaluate the prognosis and efficacy of R/R MM patients given BCMA-specific CAR-T cell therapy.