Molecular Subsets, Integrative Genomics, and Tissue Models of Scleroderma
Michael L. Whitfield, PhD
Geisel School of Medicine at Dartmouth
Michael L. Whitfield, PhD
Geisel School of Medicine at Dartmouth
Dr. Whitfield: SRF-funded work from my lab has allowed us to understand the patient-to-patient and disease-stage variability seen in scleroderma, link this variability to disease progression and identify molecular mechanisms that results in fibrosis of the skin, internal organ dysfunction (e.g.: GI symptoms), and pulmonary problems. My lab, with SRF support, has identified “molecular fingerprints” of scleroderma that determine where a patient is in their disease progression and allow us to identify drugs that may be useful in treating these patients. We have also linked these scleroderma disease states to model systems that we can use to better understand the disease. These include mouse models of disease in which we can test hypotheses, but more recently, we have developed 3-dimensional skin-like tissues that resemble human scleroderma at a molecular level. These culture models are made using scleroderma or healthy control skin cells and reproduce many disease features (skin thickness and fibrosis) that we observe in patients. Robust model systems allow us to test our hypotheses about how scleroderma progresses and what drives it—all critical parts of developing effective treatments.
A second major component of my lab is integrating the genomic data generated from my laboratory as well the vast amounts of genomic data available in the public domain, to better understand scleroderma pathogenesis. These analyses use bioinformatics, gene-gene networks, and systems biology to understand how groups of genes act together (or against each other) in scleroderma patients. We have been able to use these methods to generate a molecular model of scleroderma pathogenesis. We have been able to further show that the molecular processes that drive skin fibrosis are likely the same processes that are driving disease in other organ systems (GI tract and lungs) of the body. We are now testing these hypotheses by analyzing data from multiple organs from single patients and asking if they show the same deregulated molecular processes. We are also performing molecular experiments in model systems to test our hypotheses. These data tell us that a common mechanism is likely driving disease across organs in scleroderma patients. Our goal is to target this fundamental mechanism therapeutically.
Finally, my lab is working to actively translate our findings from bench to the bedside. These studies have included development of molecular measures of disease severity that can be used in clinical trials, diagnostic markers that identify a patient’s molecular state (i.e., which gene expression fingerprint is found in a patient), and finally, using our data to identify novel therapeutic targets and then establishing collaborative efforts to develop therapies against those targets. Our goal is to bring precision medicine efforts that are now becoming commonplace in cancer to scleroderma.
We have developed multi-tissue networks that implicate cells of the innate immune system (such as alternatively-activated macrophages and dendritic cells) that we believe are driving scleroderma in skin and internal organs affected by the disease. We have shown that these cells produce many of the molecules that have been implicated in driving scleroderma. Our network methods have also been used to perform a meta-analysis of multiple scleroderma clinical trials and we have been able to use these methods to predict possible combination therapies. We are performing experiments in mouse models and in our model skin-equivalents to confirm that eliminating these cells prevents fibrosis, something that has already been shown in other diseases such as kidney fibrosis.
Diagnostic assays that we have developed (in part with SRF funding) are showing promise in our efforts to target therapies to particular patients. In a recent example, we identified patients in the Scleroderma: Cyclophosphamide or Transplantation (SCOT) trial that were most and least likely to benefit from this lengthy and burdensome therapy. We analyzed gene expression data from the blood cells of these patients in the trial, and classified them into molecular subsets using a machine learning-based algorithm we developed over several years. We found that for one group of participants, called the ‘normal-like group,’ event-free survival did not differ between patients receiving transplant and patients receiving cyclophosphamide. In contrast, for participants in another group, called the ‘fibroproliferative group,’ the data showed a statistically significant improvement in event-free survival in patients receiving transplant as compared to patients receiving cyclophosphamide. This suggests that patients who fall into this group are more likely to benefit from stem cell transplant. This is an important finding because patients who fall into the fibroproliferative group tend not to respond to immunosuppressive therapy.
We have also implemented new efforts to develop novel therapeutics that can be used to treat scleroderma. These include efforts to target the cells driving scleroderma in collaboration with academic and industry partners. In particular, we are leveraging the methods pioneered in cancer immunotherapy at Dartmouth to develop immunotherapy for patients with scleroderma. We hope to combine our diagnostic assays and therapeutic targeting to develop a precision medicine strategy in scleroderma.
Our work is providing a comprehensive molecular mechanism for scleroderma and demonstrating that a personalized medicine approach based on patient molecular subset will help us get patients to the most effective therapy, and also guide the development of better therapies. The active translation of our work from bench to bedside has resulted in our molecular subsets being actively used in scleroderma clinical trials around the country. Our methods are helping physicians interpret the outcomes of these clinical trials and identify the patients most likely to improve on a particular treatment, as we have shown in the stem-cell transplantation trials. Our network-based methods are now being used to interpret the molecular data from clinical trials so we can understand why some treatments work and some do not; we have also used these methods to predict combinations of drugs that may be most beneficial to scleroderma patients. We hope that our efforts to find new and improved therapies will ultimately benefit patients by developing drugs (or combinations of drugs) with greater efficacy.
I would not be working on scleroderma if it were not for the Scleroderma Research Foundation, and they are part of the reason that my entire lab now works on this disease. In recent years, this research has gone beyond me as an investigator and my laboratory at the Geisel School of Medicine at Dartmouth, as students and post-doctoral fellows that trained in my lab are now studying scleroderma in their own labs, have become the computational biology experts in rheumatology and fibrotic disease, or are advocates for scleroderma in their own spheres of influence. Most of these individuals attended SRF workshops, collaborated with SRF investigators, and had their work supported by grants we receive from the SRF. With these investments, the SRF is helping to build a community of young, talented investigators to advance scleroderma research.
The interactions that the SRF has brought together some of the best scientific minds to think about this disease. In my opinion, there is no better “think tank” for scleroderma. The annual SRF Workshop provides a place and time for individuals to discuss the disease, trade their best ideas, and get expert advice from a Scientific Advisory Board that includes some of the best scientific minds in the world.