The significance of the dynamics of PET biomarkers in assessing the response to R-CHOP immunochemotherapy in patients with diffuse large B-cell lymphoma



Introduction. In recent years, the prognostic role of baseline values of 18F-FDG PET biomarkers – tumor metabolic volume and tumor glycolysis volume have been actively studied in patients with diffuse large B-cell lymphoma (DLBCL). At the same time, no researchers are studying the dynamics of these biomarkers after treatment and their relationship with the Deauville scale and response to treatment.

Objective. Assessment of the significance of changes in PET biomarkers after R-CHOP immunochemotherapy (ICT) and search for their possible relationship with the Deauville scale.

Materials and methods. An analysis of immediate clinical results (complete, partial remission, stabilization of the disease) and PET biomarkers was carried out before and after R-CHOP ICT in 46 primary patients with DLBCL. Volumetric PET biomarkers obtained after ICT were correlated with patient response to therapy and Deauville scores. The results were statistically processed using parametric and nonparametric analysis methods using the IBM SPSS Statistics v26.

Results. A pronounced direct correlation was found between the increased risk of early progression according to the criteria of the International Prognostic Index (IPI) and the levels of initial biomarkers TMTV (r=0.63, p<0.05) and TTLG (r=0.60, p< 0.05). High baseline levels of TMTV and TTLG (more than 420 cm3 and 3200 g/ml x cm3, respectively) were significantly associated with a worse immediate response to R-CHOP ICT (total OR – 3.43 and 2.87, respectively, p<0.05). Significant reductions in ΔSUVmax (>70% of baseline), ΔTTLG (>55%) and ΔTMTV (>55%) after treatment were significantly associated with a higher rate of complete clinical response. The rate of complete remission after R-CHOP ICT with a
combination of an initially low TMTV and a high average SUVmax was higher (72.7%) compared to the rates for patients
with a high TMTV and low SUVmax (10.4%). In multivariate analysis, the combination of a low baseline TMTV and a high
mean SUVmax values was an independent prognostic factor. Bivariate regression models were obtained with high multiple
correlation coefficients between the combination of IPI risk categories as a binary variable (<3 vs. ≥3) with the Deauville scale
(r = 0.812). As well as with the combination of the level of reduction in ΔSUVmax as a binary variable (<70% vs. ≥70%) with
the Deauville scale (r=0.785; p< 0.05) compared with the univariate model using only the Deauville scale (r=0.686; p<0.05).

Conclusion. For the first time, we have shown the importance of not only the basic indicators of PET biomarkers,
but also, more importantly, their dynamics and combination with the Deauville scale in assessing the immediate results
of R-CHOP ICT. The use of additional parameters increases the value of the Deauville scale, and the resulting regression
equations and models are superior to changes in the threshold values of the Deauville scale and can be used to estimate
the likelihood of achieving a complete response to R-CHOP ICT in patients with DLBCL.