Objective To explore the prognostic evaluation of the ¹⁸F-FDG PET/CT imaging response evaluation criteria in lymphoma (RECIL) for patients with non-Hodgkin's lymphoma (NHL) after autologous hematopoietic stem cell transplantation (ASCT) and compares it with the Lugano standard.

Methods The clinical data and imaging data of 86 patients with NHL diagnosed by histopathological examination in Shanxi Provincial Cancer Hospital from October 2010 to November 2021 were retrospectively analyzed, including 63 male and 23 female, aged 34.0(22.0, 47.0) years old. All patients underwent ¹⁸F-FDG PET/CT imaging before and after ASCT. According to the RECIL, all patients were evaluated for efficacy after ASCT. Based on the results of the patient efficacy evaluation, patients were divided into effective groups: complete remission (CR), partial remission (PR), and minor remission (MiR), and invalid groups: stable disease(SD), and progressive disease(PD). According to the Lugano standard, the patients were divided into complete remission group (CR), partial remission group (PR), and ineffective group (SD+PD). According to the RECIL, the patients were divided into complete remission group (CR), partial remission group (PR+MiR), andineffective group (SD+PD). The three-year overall survival (OS) period of the patients was followed up and analyzed. Pearson Chi-squared test and Mann Whitney U test were used to compare the differences in clinical features and the ¹⁸F-FDG PET/CT parameters between the effective group and the ineffective group. Univariate and multivariate Cox proportional risk regression analysis were used to screen for relevant factors affecting the prognosis of patients with NHL after ASCT. Kappa test was used to evaluate the consistency of efficacy between the RECIL and Lugano standard in evaluating the efficacy of ASCT in patients with NHL. Kaplan-Meier survival analysis was used to compare the differences in three-year OS rates between RECIL and Lugano standard in complete remission, partial remission, and ineffective groups. Log rank test was used to analyze the differences in three-year OS rates among the three groups. ROC curves were used to compare the predictiveefficacy of the RECIL and Lugano standard in three-year OS rates.

Results A statistically significant difference was observed in SUV_max (1.3(1.0, 2.0) vs. 5.2(4.8, 8.9)) between the effective group and the ineffective group after transplantation (Z=−6.149, P<0.001). The percentage of patients in the effective group who received less than two chemotherapy regimens before transplantation (65.7%(44/67)) was higher than those in the ineffective group (21.1%(4/19)), and the difference was statistically significant (χ²=11.949, P<0.001). The percentage of patients in the first-line consolidation treatment group (83.6% (56/67)) was higher than those in the ineffective group (31.6%(6/19)), and the difference was statistically significant (χ²=19.897, P<0.001). The results of univariate Cox proportional risk regression analysis showed the RECIL (HR=0.020, 95%CI: 0.003–0.155, P<0.001), post-transplant SUV_max (HR=1.177, 95%CI: 1.087–1.274, P<0.001), number of pretransplant chemotherapy regimens (HR=6.197, 95%CI: 1.338–28.711, P<0.05), and transplant timing (HR=8.808, 95%CI: 2.289–33.891, P<0.01) were prognostic factors for patients with NHL. The results of multivariate Cox proportional risk regression analysis showed the RECIL (HR=0.040, 95%CI: 0.004–0.439, P<0.01) was an independent risk factor for the prognosis of patients with NHL. The RECIL and Lugano standard were consistent in the efficacy evaluation of patients with NHL after ASCT (86.0%(74/86), Kappa=0.77, P<0.001). The Kaplan-Meier survival analysis results showed statistically significantdifferences in three-year OS rates (2.2%(1/49) vs. 0(0/18) vs. 52.0%(10/19), 2.2%(1/49) vs. 0(0/15) vs. 45.5%(10/22)) between the complete remission group, partial remission group, and ineffective group according to the RECIL and Lugano standard (χ²=42.727, 33.646; both P<0.001). The area under curve predicted by the RECIL for the three-year OS rate was slightly higher than that of the Lugano standard (0.884 vs. 0.865, Z=1.334, P>0.05).

Conclusion The RECIL can accurately evaluate the prognosis of patients with NHL after ASCT, and the RECIL and Lugano standard have similar prognostic evaluation effects on patients with NHL after ASCT.