A prognostic nomogram constructed for relapsed or refractory diffuse large B-cell lymphoma patients

Prognostic model for relapsed/refractory transplant-ineligible diffuse large B-cell lymphoma utilizing the lymphocyte-to-monocyte ratio

We conducted a multi-institutional retrospective study in 100 transplant-ineligible (TI) patients with diffuse large B-cell lymphoma (DLBCL) that relapsed or progressed after first-line R-CHOP (or -like) therapy to develop a robust predictive model for TI relapsed/refractory (r/r) DLBCL, which has a heterogeneous but poor prognosis by currently available treatment modalities other than chimeric antigen receptor T-cell (CAR-T) therapy or bispecific antibodies. The median age at relapse or progression was 76 years. The median progression-free survival (PFS) and overall survival (OS) from the first progression were 11.5 months and 21.9 months, respectively. Multivariate analysis identified low lymphocyte-to-monocyte ratio (LMR), elevated high lactate dehydrogenase, and elevated C-reactive protein at progression as independent predictors of OS. A predictive model based on these three factors, here designated as the Kyoto Prognostic Index for r/r DLBCL (KPI-R), successfully stratified their OS and PFS with statistical significance. In addition, event-free survival less than 24 months for R-CHOP and low LMR were identified as significant predictive factors for non-response in any sequence of salvage therapy. We concluded that LMR is a bonafide predictor of treatment response and prognosis in patients with TI r/r DLBCL, and may be helpful in treatment decision-making.

Development and validation of nomogram prognostic model for predicting OS in patients with diffuse large B-cell lymphoma: a cohort study in China

This study comprehensively incorporates pathological parameters and novel clinical prognostic factors from the international prognostic index (IPI) to develop a nomogram prognostic model for overall survival in patients with diffuse large B-cell lymphoma (DLBCL). The aim is to facilitate personalized treatment and management strategies. This study enrolled a total of 783 cases for analysis. LASSO regression and stepwise multivariate COX regression were employed to identify significant variables and build a nomogram model. The calibration curve, receiver operating characteristic (ROC) curve, and decision curve analysis (DCA) curve were utilized to assess the model's performance and effectiveness. Additionally, the time-dependent concordance index (C-index) and time-dependent area under the ROC curve (AUC) were computed to validate the model's stability across different time points. The study utilized 8 selected clinical features as predictors to develop a nomogram model for predicting the overall survival of DLBCL patients. The model exhibited robust generalization ability with an AUC exceeding 0.7 at 1, 3, and 5 years. The calibration curve displayed evenly distributed points on both sides of the diagonal, and the slopes of the three calibration curves were close to 1 and statistically significant, indicating high prediction accuracy of the model. Furthermore, the model demonstrated valuable clinical significance and holds the potential for widespread adoption in clinical practice. The novel prognostic model developed for DLBCL patients incorporates readily accessible clinical parameters, resulting in significantly enhanced prediction accuracy and performance. Moreover, the study's use of a continuous general cohort, as opposed to clinical trials, makes it more representative of the broader lymphoma patient population, thus increasing its applicability in routine clinical care.

A New Staging System Based on the Dynamic Prognostic Nomogram for Elderly Patients With Primary Gastrointestinal Diffuse Large B-Cell Lymphoma

Objective

The purpose of this study is to establish an accurate prognostic model based on important clinical parameters to predict the overall survival (OS) of elderly patients with primary gastrointestinal diffuse large B-cell lymphoma (EGI DLBCL).

Methods

The Cox regression analysis is based on data from the Surveillance, Epidemiology, and End Results (SEER) database.

Results

A total of 1,783 EGI DLBCL cases were eligible for the study [median (interquartile range, IQR) age, 75 (68–82) years; 974 (54.63%) males], of which 1,248 were randomly assigned to the development cohort, while 535 were into the validation cohort. A more accurate and convenient dynamic prognostic nomogram based on age, stage, radiation, and chemotherapy was developed and validated, of which the predictive performance was superior to that of the Ann Arbor staging system [C-index:0.69 (95% CI:0.67–0.71) vs. 56 (95%CI:0.54–0.58); P < 0.001]. The 3- and 5-year AUC values of ROC curves for 3-year OS and 5-year OS in the development cohort and the validation cohort were were alll above 0.7.

Conclusion

We establish and validate a more accurate and convenient dynamic prognostic nomogram for patients with EGI DLBCL, which can provide evidence for individual treatment and follow-up.

Development and Validation of a Novel Risk Stratification Model for Cancer-Specific Survival in Diffuse Large B-Cell Lymphoma

Diffuse large B-cell lymphoma (DLBCL) is a biologically and clinically heterogenous disease. Identifying more precise and individual survival prognostic models are still needed. This study aimed to develop a predictive nomogram and a web-based survival rate calculator that can dynamically predict the long-term cancer-specific survival (CSS) of DLBCL patients. A total of 3,573 eligible patients with DLBCL from 2004 to 2015 were extracted from the Surveillance, Epidemiology and End Results (SEER) database. The entire group was randomly divided into the training (n = 2,504) and validation (n = 1,069) cohorts. We identified six independent predictors for survival including age, sex, marital status, Ann Arbor stage, B symptom, and chemotherapy, which were used to construct the nomogram and the web-based survival rate calculator. The C-index of the nomogram was 0.709 (95% CI, 0.692–0.726) in the training cohort and 0.700 (95% CI, 0.671–0.729) in the validation cohort. The AUC values of the nomogram for predicting the 1-, 5-, and 10- year CSS rates ranged from 0.704 to 0.765 in both cohorts. All calibration curves revealed optimal consistency between predicted and actual survival. A risk stratification model generated based on the nomogram showed a favorable level of predictive accuracy compared with the IPI, R-IPI, and Ann Arbor stage in both cohorts according to the AUC values (training cohort: 0.715 vs 0.676, 0.652, and 0.648; validation cohort: 0.695 vs 0.692, 0.657, and 0.624) and K-M survival curves. In conclusion, we have established and validated a novel nomogram risk stratification model and a web-based survival rate calculator that can dynamically predict the long-term CSS in DLBCL, which revealed more discriminative and predictive accuracy than the IPI, R-IPI, and Ann Arbor stage in the rituximab era.

Predictive value of baseline 18F-FDG PET/CT and interim treatment response for the prognosis of patients with diffuse large B-cell lymphoma receiving R-CHOP chemotherapy

The present study aimed to investigate the prognostic value of baseline ¹⁸F-FDG PET/CT quantitative parameters and interim treatment response, and to assess whether the combination of these could improve the predictive efficacy in patients with diffuse large B-cell lymphoma (DLBCL) receiving R-CHOP chemotherapy. PET/CT images and clinical data of 64 patients with DLBCL who had undergone ¹⁸F-FDG PET/CT scan before and after 3 or 4 cycles of R-CHOP chemotherapy were retrospectively reviewed. The quantitative parameters including standardized uptake value (SUV), metabolic tumor volume (MTV), total lesion glycolysis (TLG), and maximum diameter of the maximum lesion (Dmax) were measured on baseline PET/CT images. Cox proportional hazards model was used to evaluate the influence of baseline PET/CT parameters, clinical indicators and interim treatment response on prognosis. Survival analysis was performed using Kaplan-Meier method. Receiver operating characteristic (ROC) curve analysis was performed to estimate the predictive efficacy of the combination of baseline PET/CT parameters and interim treatment response. Ann Arbor stage, International Prognostic Index (IPI), lactate dehydrogenase (LDH), necrosis, MTVmax, TLGmax, Dmax and interim treatment response showed association with 2-year progression-free survival (PFS, P<0.05). LDH, necrosis, MTVmax, MTVsum, TLGmax, TLGsum, Dmax and interim treatment response showed association with 2-year overall survival (OS, P<0.05). Ann Arbor stage, Dmax and interim treatment response were found to be independent predictors of 2-year PFS (P<0.05), while Dmax and interim treatment response were found to be independent predictors of 2-year OS (P<0.05). The PFS and OS curves of Dmax <5.7 cm group and Dmax ≥5.7 cm group, complete response (CR) group and non-CR group were significantly different, respectively (P<0.05). The baseline ¹⁸F-FDG PET/CT parameters and interim treatment response have important prognostic values in DLBCL patients who received R-CHOP chemotherapy. Combined application of Dmax and interim treatment response improved the predictive efficacy of 2-year PFS. It may be helpful to identify patients who are at high-risk of relapse and to guide early clinical intervention of these patients.

Classifying 2-year recurrence in patients with dlbcl using clinical variables with imbalanced data and machine learning methods

BACKGROUND

Treatments are limited for patients with relapsed/refractory Diffuse large B-cell lymphoma (DLBCL), and their survival rate is low. Prediction of the recurrence hazard for each patient could provide a reference regarding chemotherapy regimens for clinicians to extend patients' period of long-term remission. As current strategies cannot satisfy such need, we have established predictive models to classify patients with DLBCL with complete remission who had recurrences in 2 years from ones who did not.

METHODS

We assessed 518 patients with DLBCL and measured 52 variables of each patient. They were treated between January 2011 and July 2016. 17 variables were first selected by variable selection methods (including Lasso, Adaptive Lasso, and Elastic net). Then, we set classifiers and probability models for imbalanced data by combining the SMOTE sampling, cost-sensitive, and ensemble learning (consisting of AdaBoost, voting strategy, and Stacking) methods with the machine learning methods (Support Vector Machine, BackPropagation Artificial Neural Network, Random Forest), respectively. Last, assessed their performance.

RESULTS

The disease stage and other 5 variables are significant indicators for recurrence. The SVM with AdaBoost ensemble learning method modeling by SMOTE data performs the best (Sensitivity=97.3%, AUC=96%, RMSE=19.6%, G-mean=96%) in all classifiers. The SVM with AdaBoost method(AUC=98.7%, RMSE=17.7%, MXE=12.7%, Cal mean=3.2%, BS0=2.5%, BS1=4%, BSALL=3.1%) and random forest (AUC=99.5%, RMSE=19.8%, MXE=16.2%, Cal mean=9.1%, BS0=4.8%, BS1=2.9%, BSALL=3.9%) both modeling by SMOTE sampling data perform well in probability models.

CONCLUSIONS

This predictive model has high accuracy for almost all DLBCL patients and the six indicators can be recurrence signals.

Prognostic Nomogram and Predictive Factors in Refractory or Relapsed Diffuse Large B-Cell Lymphoma Patients Failing Front-Line R-CHOP Regimens

Background: The clinical course of relapsed or refractory (r/r) diffuse large B-cell lymphoma (DLBCL) is variable, with limited efficacy data of second-line treatment in a post-rituximab real-world context. Hence, we explored the predictors and constructed a nomogram for risk stratification in this population.

Patients and methods: Among 296 r/r DLBCL patients pretreated with R-CHOP (rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone) at the Cancer Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College between 2006 and 2017, 231 were included for nomogram construction. After randomization, we constructed the prognostic nomogram in the primary cohort (n=161) based on a multivariate analysis and confirmed it in the validation cohort (n=70). Additionally, we explored predictive factors for second-line therapy using a ordinal regression analysis.

Results: Four independent prognostic factors including rituximab in the second-line setting, initial Eastern Cooperative Oncology Group (ECOG) performance status (PS), response to front-line treatment, and invasion on progression/recurrence were used to construct the nomogram. The nomogram had a C index of 0.70 with AUC values of 0.73 and 0.71 for the primary and validation cohorts, respectively. Subsequently, three risk groups (low, intermediate, and high) were determined with median overall survival (OS) of 38.0, 25.0, and 7.0 months, respectively. Additionally, we conducted a multivariate ordinal regression analysis and identified prior response to front-line treatment (odds ratio=4.50, 95% CI: 1.84-11.27, p=0.001) and bulky disease at diagnosis (odds ratio=0.36, 95% CI: 0.182-0.702, p=0.003) were two independent factors for second-line treatment efficacy.

Conclusions: The established predictors for treatment efficacy and the novel nomogram for survival in r/r DLBCL patients could potentially be applied for risk stratification and treatment guidance in the post-rituximab era.