Development of a Novel Prognostic Score for Relapsed/Refractory Mature T-Cell and NK-Cell Lymphomas (PIRT): Results from a Global Peripheral T-Cell Lymphoma (PETAL) Consortium

The variability in global access to medications and treatment practices make it challenging to evaluate the benefits of contemporary therapies for patients with relapsed and refractory (R/R) mature T-cell and NK-cell lymphomas (MTCL and MNKCL). While numerous prognostic factors and models have been suggested to categorize the risk levels of patients with MTCL and MNKCL, they have primarily been developed for newly diagnosed patients. On the other hand, for the common subtypes of R/R MTCL and MNKCL patients, the identification of risk factors and subgroups that lead to variability in treatment outcomes and overall survival (OS) is still not well established. Thus, a new prognostic model is needed to further refine subgroups with varied outcomes in R/R patients. We conducted an international retrospective cohort study of 925 patients with R/R MTCL and MNKCL diagnosed between Jan 2010 and Sept 2021 who received either “novel” single agent (SA) or cytotoxic chemotherapy (CC) for second line treatment from 13 institutions representing 10 countries. We first split our global cohort of 763 patients with second line treatment details into a training set (80%) and a testing set (20%). Among the 21 available demographics, histological, laboratory, radiologic non-treatment characteristics, we identified 11 variables on univariable analysis to be associated with inferior OS from the start of second-line treatment. Based on the step-by-step selection in multivariable Cox regression (MVC) and clinical considerations, our final model that combined 6 of the 11 features was chosen based on the highest Concordance index (C-index) among the models that incorporated covariates retaining an independent prognostic value (p £ 0.05 in the testing set. The six variables in the final model included age being greater than 60, primary refractory disease in contrast to relapsed, histological subtypes other than AITL, extranodal sites greater than 1, Ki-67 proliferation index (%) being 40 or greater, and absolute lymphocyte count around diagnosis being lower than Lower Limit of Normal (LLN). We then created a new prognostic index for R/R TCL (PIRT) score to estimate OS from start of second-line treatment, which ranged from 0 to 6. We assigned a score of 1 to each unfavorable feature since the relative risks associated with each of these 6 factors are similar. When patients were initially categorized into 6 subgroups, we observed a clear declining trend of OS with an increasing score. Patients were stratified into 3 final risk group: low- (0-1 risk factor), intermediate- (2-3 risk factors), or high- (3-4 risk factors) groups, which showed notable similarity between their predicted and actual survival. In the training set, the median 3-yr OS for the low-risk group was 75.2% (95% CI: 29.7 - 85.2) vs intermediate-risk group (50.6%, 95% CI: 37.7 - 58.9) vs the high-risk group (16%, 95% CI: 7.7 - 26.9). The three risk groups were also segregated in the testing set with disparate 3-yr OS with low-risk (57.1%, 95% CI: 17.1 - 83.7) vs intermediate-risk (23.3%, 95% CI: 8.7 - 41.9) vs the high-risk (7.0%, 95% CI: 0.4 - 26.9). An analysis of 1000 bootstrapped testing sets showed that our PIRT score achieved a higher prediction probability with an average C-index of 0.7 (95% CI: 0.63 - 0.76) compared to IPI (0.56, 95% CI: 0.5 - 0.63) and PIT (0.59, 95% CI: 0.52 - 0.65). The paired t-tests between IPI vs. PIRT and PIT vs. PIRT scores confirmed the statistical significance of this performance difference. We also validated our prognostic model's survival prediction power in an independent multicenter study cohort and developed a web-based calculator for healthcare professionals to use: Website for PIRT Score Calculator In summary, we developed a new prognostic scoring system for patients with R/R MTCL and MNKCL that was validated in training, testing and independent datasets that has enabled their segregation and has the potential to facilitate clinical decision.

Authors

Institutions

Publication Details

Journal
Blood
Published
2024-11-05
DOI
https://doi.org/10.1182/blood-2024-201874
Primary Topic
Lymphoma Diagnosis and Treatment
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Development of a Novel Prognostic Score for Relapsed/Refractory Mature T-Cell and NK-Cell Lymphomas (PIRT): Results from a Global Peripheral T-Cell Lymphoma (PETAL) Consortium

Owen A. O’Connor, Mark N Sorial, Noriaki Yoshida, Mubarak M. Al-Mansour et al.
Blood
Lymphoma Diagnosis and Treatment
article

Development of a Novel Prognostic Score for Relapsed/Refractory Mature T-Cell and NK-Cell Lymphomas (PIRT): Results from a Global Peripheral T-Cell Lymphoma (PETAL) Consortium

Owen A. O’Connor, Mark N Sorial, Noriaki Yoshida, Mubarak M. Al-Mansour, Ahmed Sawas, Marianna Gentilini, Ijeoma Julie Eche, Josie Ford, Govind Bhagat, Tinatin Muradashvili, Monica Civallero, Caroline MacVicar, Ramzi Hamouche, K. Kariya, Jeffrey A. Barnes, Martina Manni, Devavrat Shah, Henry Miles Prince, Sridhar Epari, Massimo Federico, Kusha Chopra, Helen Ma, Alexandra W Lenart, Alok Shetty, Makoto Iwasaki, Mwanasha H. Merrill, Eliana Miranda, Jessy Xinyi Han, Jackielyn Cuenca Alturas, Min Ji Koh, A. DiSciullo, Estelle Verburgh, Carrie van der Weyden, Changyu Shen, Yu Ri Kim, Amy Ku, Sang Eun Yoon, Beatrice Casadei, Won Seog Kim, Takeshi Okatani, Leora S. Boussi, María Elena Cabrera, Salvia Jain, Min Jung Koh, Aditya Garg, Eric D. Jacobsen, Girisha Panchoo, Luke Peng, Robert N. Stuver, Tetiana Skrypets, Thomas Eipe, Shambhavi Singh, Francine M. Foss, Zainab Mohamed, Sarah McCue Horwitz, Pier Luigi Zinzani, Judith Gabler, Jin Seok Kim, Enrica Marchi, Y. Choi, Carlos Chiattone, Hasmukh Jain, Manju Sengar, Tanuja Shet, Saswata Saha, Sean M McCabe, Emmanuel Nwodo, Forum Bhanushali, Maria J. J. Fernandez Turizo
article en

Abstract

The variability in global access to medications and treatment practices make it challenging to evaluate the benefits of contemporary therapies for patients with relapsed and refractory (R/R) mature T-cell and NK-cell lymphomas (MTCL and MNKCL). While numerous prognostic factors and models have been suggested to categorize the risk levels of patients with MTCL and MNKCL, they have primarily been developed for newly diagnosed patients. On the other hand, for the common subtypes of R/R MTCL and MNKCL patients, the identification of risk factors and subgroups that lead to variability in treatment outcomes and overall survival (OS) is still not well established. Thus, a new prognostic model is needed to further refine subgroups with varied outcomes in R/R patients. We conducted an international retrospective cohort study of 925 patients with R/R MTCL and MNKCL diagnosed between Jan 2010 and Sept 2021 who received either “novel” single agent (SA) or cytotoxic chemotherapy (CC) for second line treatment from 13 institutions representing 10 countries. We first split our global cohort of 763 patients with second line treatment details into a training set (80%) and a testing set (20%). Among the 21 available demographics, histological, laboratory, radiologic non-treatment characteristics, we identified 11 variables on univariable analysis to be associated with inferior OS from the start of second-line treatment. Based on the step-by-step selection in multivariable Cox regression (MVC) and clinical considerations, our final model that combined 6 of the 11 features was chosen based on the highest Concordance index (C-index) among the models that incorporated covariates retaining an independent prognostic value (p £ 0.05 in the testing set. The six variables in the final model included age being greater than 60, primary refractory disease in contrast to relapsed, histological subtypes other than AITL, extranodal sites greater than 1, Ki-67 proliferation index (%) being 40 or greater, and absolute lymphocyte count around diagnosis being lower than Lower Limit of Normal (LLN). We then created a new prognostic index for R/R TCL (PIRT) score to estimate OS from start of second-line treatment, which ranged from 0 to 6. We assigned a score of 1 to each unfavorable feature since the relative risks associated with each of these 6 factors are similar. When patients were initially categorized into 6 subgroups, we observed a clear declining trend of OS with an increasing score. Patients were stratified into 3 final risk group: low- (0-1 risk factor), intermediate- (2-3 risk factors), or high- (3-4 risk factors) groups, which showed notable similarity between their predicted and actual survival. In the training set, the median 3-yr OS for the low-risk group was 75.2% (95% CI: 29.7 - 85.2) vs intermediate-risk group (50.6%, 95% CI: 37.7 - 58.9) vs the high-risk group (16%, 95% CI: 7.7 - 26.9). The three risk groups were also segregated in the testing set with disparate 3-yr OS with low-risk (57.1%, 95% CI: 17.1 - 83.7) vs intermediate-risk (23.3%, 95% CI: 8.7 - 41.9) vs the high-risk (7.0%, 95% CI: 0.4 - 26.9). An analysis of 1000 bootstrapped testing sets showed that our PIRT score achieved a higher prediction probability with an average C-index of 0.7 (95% CI: 0.63 - 0.76) compared to IPI (0.56, 95% CI: 0.5 - 0.63) and PIT (0.59, 95% CI: 0.52 - 0.65). The paired t-tests between IPI vs. PIRT and PIT vs. PIRT scores confirmed the statistical significance of this performance difference. We also validated our prognostic model's survival prediction power in an independent multicenter study cohort and developed a web-based calculator for healthcare professionals to use: Website for PIRT Score Calculator In summary, we developed a new prognostic scoring system for patients with R/R MTCL and MNKCL that was validated in training, testing and independent datasets that has enabled their segregation and has the potential to facilitate clinical decision.

BloodVol. 144(Supplement 1)
University of Modena and Reggio Emilia (IT), Brigham and Women's Hospital (US), Northeastern University (US), Biogen (United States) (US), Beth Israel Deaconess Medical Center (US), Memorial Sloan Kettering Cancer Center (US), Harvard University (US), University of Cape Town (ZA), Universidade Estadual de Campinas (UNICAMP) (BR), Georgetown University (US), Yonsei University (KR), University of California, Irvine (US), Radiation Effects Research Foundation (JP), Columbia University Irving Medical Center (US), Severance Hospital (KR), Tata Memorial Hospital (IN), Peter MacCallum Cancer Centre (AU), Samsung Medical Center (KR), Groote Schuur Hospital (ZA), King Abdulaziz Medical City (SA), Yale Cancer Center (US), Yale University (US), Massachusetts General Hospital (US), Hospital del Salvador (CL), Hiroshima Red Cross Hospital & Atomic-bomb Survivors Hospital (JP), Dana-Farber Cancer Institute (US), Istituto di Ematologia di Bologna (IT), Istituto Tumori Bari (IT), Gangnam Severance Hospital (KR), Istituti di Ricovero e Cura a Carattere Scientifico (IT), Yonsei University Health System (KR), University of Virginia (US), Massachusetts Institute of Technology (US), Columbia University (US), Sungkyunkwan University (KR), University of Bologna (IT)
Partnerships for the goals
Openalex Percentile: Top 64%
Lymphoma Diagnosis and Treatment
DOISourceOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.