Steven Rhodes, M.D., Ph.D.
1044 W. Walnut Street
Indianapolis, IN 46202
Phone: (317) 278-9290
Research Program Membership
Department of Pediatrics
Division of Hematology/Oncology
IU School of Medicine
Dr. Rhodes's research interests include:
My research focus is centered on driving the preclinical and early-phase clinical translation of new therapies to treat and prevent the development of neurofibromatosis type 1 (NF1)-associated malignant peripheral nerve sheath tumors (MPNST). MPNST is a devastating form of sarcoma that represents the leading cause of death in persons with NF1. It has one of the highest death rates among any sarcoma with an overall survival between 20-50% and is refractory to conventional cytotoxic chemotherapy and radiation (Kim et al., 2017). The majority of MPNST arise from pre-existing benign plexiform neurofibromas (PNF) and precancerous atypical neurofibromas (ANF) suggesting that these precursor tumors themselves may encode important information critical to unlocking the molecular origins of MPNST. Notably a high proportion of atypical neurofibroma and MPNST exhibit loss of the 9p21.3 cytoband encoding the INK4/ARF (CDKN2A/B) tumor suppressor locus (Beert et al., 2011; Pemov et al., 2019), which specifies 3 pivotal tumor suppressors: P16INK4A, P15INK4B, AND P14ARF (P19ARF in mice) that collectively regulate the activities of the retinoblastoma (RB) protein and P53 transcription factor and play a critical role in triggering incipient cancer cells to enter a growth arrest called senescence following oncogenic insult. During my post-doctoral research in the laboratory of Dr. Wade Clapp, I demonstrated using genetically engineered mouse models that superimposed loss of the Cdkn2a alternate reading frame (Arf, P19ARF) is sufficient to promote the progression of PNF to precancerous atypical neurofibromas by allowing tumorigenic Nf1 deficient Schwann cells to escape senescence (Rhodes et al., 2019). These Nf1-Arf mutant mice developed tumors histopathologically indistinguishable from human atypical neurofibroma, which spontaneously progressed to malignant peripheral nerve sheath tumor with high penetrance. We have since developed models in which multiple components of the Ink4/Arf locus are deleted, alone and in combination which demonstrate similar but distinctive phenotypes (Rhodes et al., manuscript in preparation). My future research plan will capitalize on these models, which share the same mutational signatures as NF1 patients and provide a valuable window of insight into the natural history of the disease, the molecular mechanisms underpinning tumor progression, and further, will allow for preclinical evaluation of new therapeutic strategies designed not only to treat MPNST but to prevent plexiform and atypical neurofibroma precursors from undergoing malignant transformation. Research Plan: Over the next 5 years I plan to continue this work in the novel directions described below: Direction 1: Leveraging clinical tissue samples, GEM models, transcriptome/kinome profiling, and single cell biology to understand the molecular mechanisms underpinning malignant transformation of plexiform and atypical neurofibromas to MPNST. Funding: Francis S. Collins Scholarship in Neurofibromatosis Clinical and Translation Research from the Neurofibromatosis Therapeutics Acceleration Program (NTAP) will provide full salary support for my NF1 related clinical and research activities through June, 2023 (PI: Rhodes) High mobility group protein A2 (HMGA2) is a chromatin architectural modifying protein that broadly modulates gene transcription by altering chromatin structure and recruiting protein complexes critical for promoter transactivation. HMGA2 is silenced in adult somatic tissues by lethal 7 (Let7) micro-RNAs but becomes aberrantly reactivated in cancer (Hammond & Sharpless, 2008). RNA sequencing of MPNSTs, plexiform neurofibroma and ANNUBP precursors arising in our GEM models revealed profound deregulation of the Let7-HMGA2 axis as PNF and ANF undergo malignant transformation. Immunochemical evaluation has confirmed these findings in both murine and human NF1 associated MPNSTs. We further found that shRNA knockdown of HMGA2 strongly inhibited the growth of MPNST cancer stem cells and cell lines in vitro. We therefore hypothesize that the deregulation of the Let7-HMGA2 axis reprograms broad transcriptional and kinase networks to drive MPNST progression by reactivating programs of embryotic growth in their tumor initiating cells of origin. We propose to investigate this hypothesis using a series of innovative experimental techniques. Building upon an approach pioneered by our collaborator, Dr. Lu Le, we have isolated the tumorigenic cells of origin for nerve sheath derived tumors and shown that deletion of both Nf1 and Ink4a/Arf and reimplantation into the sciatic nerve, these cells form tumors histopathologically indistinguishable from human plexiform neurofibroma and ANNUBP and then spontaneously progress to MPNST (Rhodes, manuscript in preparation). We will leverage this model to determine the molecular mechanisms by which Let7 and HMGA2 reprogram the kinome (collaborator: Dr. Steve Angus, IUSM) and transcriptome of this unique population of cancer stem cells using shRNA lentiviral techniques to modulate HMGA2 and Let7 expression. By re-implanting these cells in vivo, we will further establish whether HMGA2 is required for malignant transformation of plexiform and atypical neurofibroma precursors to MPNST. This experimental approach will fill critical knowledge gaps by defining Let7 and HMGA2 target genes and kinases that cooperate with loss of the Nf1 and Ink4a/Arf tumor suppressor loci to engender malignant transformation of plexiform and atypical neurofibroma in an orphan cancer predisposition syndrome with no effective pharmacotherapies. If proven correct, we will have identified a targetable pathway to treat or even prevent the occurrence of these devastating tumors in persons with NF1. Regardless, these studies will provide a platform to accelerate the translation of new therapeutic modalities to treat, forestall and ultimately prevent MPNST progression. Direction 2: To identify and validate molecular diagnostic signatures to prospectively predict progression of plexiform and atypical neurofibroma to MPNST and guide clinical decision making for early intervention. Funding: Pediatric Scientist Development Award (K12) from the (PI: Rhodes), Developmental and Hyperactive Ras Tumor SPORE (PI: Clapp), Precision Health Initiative (PI: Foroud) Plexiform and atypical neurofibroma proceed along diverse evolutionary trajectories. While the majority remain indolent for years or even decades, a subset will undergo malignant transformation. The lifetime incidence of progression to MPNST is between 8-16%, yet this often occurs with little to no early warning signs. While exome sequencing has improved our understanding of the genomic events such as INK4/ARF (CDKN2A/B) loss driving tumor progression along the plexiform neurofibroma to MPNST continuum (Beert et al., 2011; Pemov et al., 2019), the transcriptional and microenvironmental alterations that promote this evolution are largely uncharted. Presently there are no molecular diagnostic tests to forecast the biological behavior of plexiform or atypical neurofibromas at risk for undergoing malignant transformation. Hence clinicians and NF1 patients alike are faced with the difficult dilemma of how to monitor these lesions and when to consider potentially highly morbid surgical intervention at an early stage and preempt the development of MPNST. RNA sequencing provides an opportunity to characterize transcriptional changes underpinning early events in disease progression, however only a small number of atypical neurofibroma transcriptomes have been published to date due to the paucity of high-quality flash frozen tissues. Further, intratumoral heterogeneity presents a distinct challenge to understanding the molecular drivers of tumor progression in flash frozen tissues. Peripheral nerve sheath tumors in NF1 patients are heterogenous neoplasms, and varying grades along the PNF-MPNST continuum are frequently observed in a biopsy specimen. To overcome these barriers, I have employed a highly multiplexable hybridization-based approach to quantify gene expression in spatially restricted tissue regions across 35 NF1-associated peripheral nerve sheath tumor samples. Through multi-regional sampling, I have begun to identify alterations in signal transduction, immune response, and stromal factors within the tumor microenvironment associated with disease progression. Further, we follow these changes longitudinally within individual subjects who developed progressive atypical neurofibromas and/or MPNST, providing a unique window of insight into the nascent events of peripheral nerve sheath tumor evolution. Through a K08 award submission, I propose to interrogate these changes through single-cell analytics, integration with the GEM models described above, and ultimately, a prospective R01-funded trial to validate a molecular diagnostic signature to identify plexiform and atypical neurofibromas that at are high risk for undergoing malignant transformation and guide clinical decision making for early intervention. Direction 3: Develop chemotherapeutic approaches to treat and ultimately prevent MPNST from occurring. Funding: Francis S. Collins Scholarship in Neurofibromatosis Clinical and Translation Research from NTAP (PI: Rhodes), IU Simon Cancer Center EDT/TMM Shark Tank (PI: Angus/Clapp/Rhodes). Beyond the lack of molecular biomarkers to identify the subset of plexiform and atypical neurofibromas that are likely to exhibit aggressive biological behavior, another challenge facing NF1 patients and clinicians alike is the fact that there are presently no effective chemotherapeutic agents that can alter the natural history of the disease and prevent the malignant transformation of these precancerous tumors. Surgery remains the standard of care, but is often highly morbid or altogether infeasible due to the high degree of vascularity and entanglement of these nerve sheath derived tumors with other vital structures. Using our GEM models and our ability to isolate and re-implant the cells of origin for these tumors, recapitulating their stepwise progression in vivo, we have begun to interrogate therapeutic strategies for MPNST chemoprevention. As a proof of concept, we have shown that the MDM2 inhibitor, RG7388, which blocks the P53 dependent functions of p19ARF, can not only markedly reduce the volume of MPNST but can in a subset of mice, delay or prevent the development of MPNST from pre-existing plexiform and atypical neurofibroma (Rhodes, manuscript in preparation). In parallel experiments, we will also test the impact of the histone deacetylase inhibitor, panobinostat, which we have shown epigenetically modulates HMGA2 expression in forestalling malignant transformation. Collectively, these studies will provide unparalleled opportunities to interrogate the molecular mechanisms underlying NF1 associated nerve sheath tumor progression and to identify new and effective therapies for MPNST treatment and prevention where none exist today. Furthermore, understanding response heterogeneity and resistance mechanisms through single transcriptomics and kinome profiling (in collaboration with Dr. Angus) will provide unique opportunities for mechanism-based studies evaluating combination therapies through DOD and bench-to-bedside R01 support to accelerate the advancement of promising candidates to early phase clinical trials. Summary and Long-Term Goals: My background in GEM models, molecular biology and additional expertise in applied computational methods to analyze next generation sequencing and proteomics data acquired during my fellowship training has positioned me well to address these research questions in my independent scientific pursuits. Importantly, the research directions outlined here, I believe, have the potential to make a true impact on improving the survival, reducing suffering, and alleviating the burden of care for persons with NF1 who are affected by these devastating tumors. The studies proposed here will provide critical insight into the mechanisms underpinning malignant transformation of plexiform and atypical neurofibromas to MPNST, facilitate discovery of molecular biomarkers to guide early therapeutic interventions by identifying patients at risk, and will accelerate the development of novel chemotherapeutic strategies to delay and or halt malignant transformation. The close integration between the NF multidisciplinary clinic and my research focus will allow for seamless and expedient translation of these findings from bench-to-bedside and further establish IUSM/IUSCC as a Center of Excellence, offering state-of-the art, personalized care for persons affected by NF.
Fellowship - Indiana University School of Medicine, Indianapolis, IN 06/2021
Residency - Indiana University School of Medicine, Indianapolis, IN 06/2017
Ph.D. - Indiana University School of Medicine, Indianapolis, IN 06/2013
M.D. - Indiana University School of Medicine, Indianapolis, IN 05/2015