Shelli R Kesler1,2,3, Melissa L Petersen4, Vikram Rao5,6, Rebecca A Harrison7, Oxana Palesh8,9. 1. Cancer Neuroscience Laboratory, School of Nursing, University of Texas at Austin, 1710 Red River St, Austin, TX, 78712, USA. srkesler@austin.utexas.edu. 2. Department of Diagnostic Medicine, Dell School of Medicine, University of Texas at Austin, Austin, TX, USA. srkesler@austin.utexas.edu. 3. LIVESTRONG Cancer Institutes, Dell School of Medicine, University of Texas at Austin, Austin, TX, USA. srkesler@austin.utexas.edu. 4. Department of Family Medicine, University of North Texas Health Science Center, Fort Worth, TX, USA. 5. Cancer Neuroscience Laboratory, School of Nursing, University of Texas at Austin, 1710 Red River St, Austin, TX, 78712, USA. 6. Department of Diagnostic Medicine, Dell School of Medicine, University of Texas at Austin, Austin, TX, USA. 7. Department of Neuro-Oncology, University of Texas MD Anderson Cancer Center, Houston, TX, USA. 8. Department of Psychiatry and Behavioral Sciences, Stanford University School of Medicine, Palo Alto, CA, USA. 9. Stanford Cancer Institute, Palo Alto, CA, USA.
Abstract
PURPOSE: Cancer-related cognitive impairment (CRCI) is a common neurotoxicity among patients with breast and other cancers. Neuroimaging studies have demonstrated measurable biomarkers of CRCI but have largely neglected the potential heterogeneity of the syndrome. METHODS: We used retrospective functional MRI data from 80 chemotherapy-treated breast cancer survivors to examine neurophysiologic subtypes or "biotypes" of CRCI. The breast cancer group consisted of training (N = 57) and validation (N = 23) samples. RESULTS: An unsupervised clustering approach using connectomes from the training sample identified three distinct biotypes. Cognitive performance (p < 0.05, corrected) and regional connectome organization (p < 0.001, corrected) differed significantly between the biotypes and also from 103 healthy female controls. We then built a random forest classifier using connectome features to distinguish between the biotypes (accuracy = 91%) and applied this to the validation sample to predict biotype assignment. Cognitive performance (p < 0.05, corrected) and regional connectome organization (p < 0.005, corrected) differed significantly between the predicted biotypes and healthy controls. Biotypes were also characterized by divergent clinical and demographic factors as well as patient reported outcomes. CONCLUSIONS: Neurophysiologic biotypes may help characterize the heterogeneity associated with CRCI in a data-driven manner based on neuroimaging biomarkers. IMPLICATIONS FOR CANCER SURVIVORS: Our novel findings provide a foundation for detecting potential risk and resilience factors that warrant further study. With further investigation, biotypes might be used to personalize assessments of and interventions for CRCI.
PURPOSE:Cancer-related cognitive impairment (CRCI) is a common neurotoxicity among patients with breast and other cancers. Neuroimaging studies have demonstrated measurable biomarkers of CRCI but have largely neglected the potential heterogeneity of the syndrome. METHODS: We used retrospective functional MRI data from 80 chemotherapy-treated breast cancer survivors to examine neurophysiologic subtypes or "biotypes" of CRCI. The breast cancer group consisted of training (N = 57) and validation (N = 23) samples. RESULTS: An unsupervised clustering approach using connectomes from the training sample identified three distinct biotypes. Cognitive performance (p < 0.05, corrected) and regional connectome organization (p < 0.001, corrected) differed significantly between the biotypes and also from 103 healthy female controls. We then built a random forest classifier using connectome features to distinguish between the biotypes (accuracy = 91%) and applied this to the validation sample to predict biotype assignment. Cognitive performance (p < 0.05, corrected) and regional connectome organization (p < 0.005, corrected) differed significantly between the predicted biotypes and healthy controls. Biotypes were also characterized by divergent clinical and demographic factors as well as patient reported outcomes. CONCLUSIONS: Neurophysiologic biotypes may help characterize the heterogeneity associated with CRCI in a data-driven manner based on neuroimaging biomarkers. IMPLICATIONS FOR CANCER SURVIVORS: Our novel findings provide a foundation for detecting potential risk and resilience factors that warrant further study. With further investigation, biotypes might be used to personalize assessments of and interventions for CRCI.
Entities:
Keywords:
Breast Cancer; Cognition; Connectome; MRI; Machine Learning
Authors: Iman Sahnoune; Taeko Inoue; Shelli R Kesler; Shaefali P Rodgers; Omaima M Sabek; Steen E Pedersen; Janice A Zawaski; Katharine H Nelson; M Douglas Ris; J Leigh Leasure; M Waleed Gaber Journal: Neuro Oncol Date: 2018-04-09 Impact factor: 12.300
Authors: Ali Amidi; S M Hadi Hosseini; Alexander Leemans; Shelli R Kesler; Mads Agerbæk; Lisa M Wu; Robert Zachariae Journal: J Natl Cancer Inst Date: 2017-12-01 Impact factor: 13.506
Authors: Shelli R Kesler; Robert Ogg; Wilburn E Reddick; Nicholas Phillips; Matthew Scoggins; John O Glass; Yin Ting Cheung; Ching-Hon Pui; Leslie L Robison; Melissa M Hudson; Kevin R Krull Journal: Brain Connect Date: 2018-08
Authors: Shelli R Kesler; Tien Tang; Ashley M Henneghan; Michelle Wright; M Waleed Gaber; Oxana Palesh Journal: Front Neurol Date: 2021-10-29 Impact factor: 4.003
Authors: Tien T Tang; Janice A Zawaski; Shelli Kesler; Christine A Beamish; Taeko Inoue; Emma C Perez; Lawrence Bronk; Falk Poenisch; Tina M Briere; Omaima M Sabek; David R Grosshans; M Waleed Gaber Journal: Int J Radiat Oncol Biol Phys Date: 2021-09-09 Impact factor: 7.038