Scientists at Stanford University School of Medicine have found that statins, a commonly prescribed type of cholesterol-lowering drug, may also be an effective, if unexpected, treatment for the inflammatory bowel disease, ulcerative colitis (UC). By tapping into publicly available datasets of anonymized patient health information, including transcriptomic data from ulcerative colitis samples, the team identified a genomic signature associated with UC disease, and then identified the FDA-approved statin, atorvastatin, as a potential candidate for reducing UC symptoms, based on that genomic signature. When they analyzed anonymized UC patient records they found that patients who were taking statins were significantly less likely to need surgery or to be hospitalized.
“At this point, one could argue that this data shows a strong enough connection to start prescribing statins for ulcerative colitis,” said Purvesh Khatri, PhD, associate professor of medicine and of biomedical data science, who led the research. “I think we’re almost there. We need to validate the effects a bit more stringently before moving it into the clinic.”
Khatri and colleagues reported on their findings in the Journal of the American Medical Informatics Association, in a paper titled, “Computational drug repositioning of atorvastatin for ulcerative colitis.” They suggest that their study approach could also represent a systematic framework for marrying molecular and clinical data, with a view to repurposing FDA-approved drugs for a wide range of diseases. Graduate students Lawrence Bai and Madeline Scott are co-lead authors of the published report.
UC is an inflammatory bowel disease that affects about a million people in the United States, and the incidence and prevalence of the disorder are on the rise worldwide, the authors commented. The disorder causes inflammation and ulcers in the bowel, leaving patients vulnerable to symptoms including abdominal pain, blood in the stool, constipation, and fatigue.
Currently, the only lines of defense against UC are anti-inflammatory drugs, which don’t always work, and colectomy, which is the surgical removal of part or all of the colon. “The undesirability of colectomies underscores the need for additional medication options for patients with UC that can reduce colectomy rates,” the team wrote. “About 30% of ulcerative colitis patients eventually have to undergo a colectomy as a last resort,” added Khatri. “It’s a drastic measure; you’re removing part of your body. So we thought, ‘Can we use available data to see whether drugs that are already approved by the FDA can be repurposed to better treat these patients?’”
Khatri and his team began their research by analyzing publicly available genomic data from hundreds of patients with ulcerative colitis who had undergone a colon biopsy, a relatively common practice that helps doctors to diagnose the disease and its severity. Specifically, Khatri and his team were looking for genomic signatures, or patterns of gene activity, that seemed to persist in most patients with the condition. “We chose to integrate multiple independent datasets that collectively represent biological, clinical, and technical heterogeneity observed in the real-world patient population to identify a robust gene signature for UC,” the authors explained. “We performed a multi-cohort analysis of 272 colon biopsy transcriptome samples across 11 publicly available datasets to identify a robust UC disease gene signature.” Khatri further commented, “We looked at national and international data, and we found a disease signature that was robust across all the datasets irrespective of whether the patient was experiencing a flare in disease.”
From there, it was a matter of identifying how certain drugs affected the gene activity associated with UC. Khatri turned to data from previously conducted lab studies in cells that showed how certain drugs changed the activity of genes. The idea was to find the drugs that seemed to reverse the gene signature associated with ulcerative colitis. For instance, if patients with UC had a dip in the activity of gene A and B, the team looked for drugs that increased activity in those genes. They looked only at FDA-approved drugs so that, if they found one that worked, it could be rolled out to patients sooner. “We hypothesized that FDA-approved drugs with transcriptome profiles inversely correlated with our UC signature could reduce disease pathology,” the team noted. “To test this hypothesis, we correlated the UC signature with transcriptome profiles of one of 781 FDA-approved small molecules … We used transcriptome profiles from Library of Integrated Network-Based Cellular Signatures (LINCS) L1000 platform to compare drug signatures against our UC signature.”
After cross-referencing the genomic and experimental sets of data, the team identified three drugs—the anticancer drugs vemurafenib and gefitinib, and the statin atorvastatin—which effectively reversed the gene signature of ulcerative colitis. “The first two were chemotherapy drugs, which of course you wouldn’t prescribe to someone due to serious side effects, but the third was a statin,” Khatri said. “Statins are generally safe enough that some doctors joke they should be put in the water.”
The next step would usually be to set up a clinical trial, but Khatri took a different approach. Statins are among the most commonly prescribed drugs in the United States, so it seemed reasonable to think that a relatively large number of patients with ulcerative colitis might also be taking statins to help manage their cholesterol. So, instead of turning to a clinical trial, the team turned to data from electronic health records. “We examined outcomes in patients with UC from Stanford University’s STARR electronic health records database and the Optum Clinformatics DataMart healthcare claims database,” the investigators explained. “The final study cohorts included 827 subjects in STARR (596 in the comparator group; 231 in the atorvastatin group), and 7821 subjects in Optum (4940 in the comparator group; 2881 in the atorvastatin group).” Khatri noted, “We were able to see if they had ulcerative colitis, if they were on statins and whether they’d needed a colectomy.”
The data showed that people with ulcerative colitis who were taking statins, regardless of their age, had about a 50% decrease in colectomy rates and were less likely to be hospitalized. In addition, ulcerative colitis patients who were taking statins were prescribed other anti-inflammatory medications at a lower rate.
While it’s not entirely known how statins might reduce symptoms of the disease, Khatri pointed out that they are known to have some sort of general anti-inflammatory capacity. “We hypothesize that although atorvastatin may not work as a short-term alternative to steroid to combat acute inflammatory flares, it may have potential long-term benefits,” the scientists stated. “Indeed, we show that long-term atorvastatin use conferred increased protection compared to short-term atorvastatin use … and this effect has been noted in atorvastatin suppression of other autoinflammatory diseases.”
And while the exact mechanism of action of the anti-inflammatory properties of atorvastatin are not well defined, the team continued, “… many groups have provided in vitro and in vivo evidence pointing to a variety of pathways in which atorvastatin may derive its pleiotropic properties, including TNF, CXCL10, and MCP-1/CCL2.”
The researchers further noted that because atorvastatin is already recognized as a safe and effective drug for treating cardiovascular disease in humans, and the side effect profile is milder compared to many other current UC drugs, their results support further investigations into the use of the drug for treating patients with UC. “Prospective controlled clinical trials are needed to confirm whether atorvastatin treatment would benefit patients with UC,” they stated.
The other benefit of the study, Khatri said, is that it provides a framework for how to identify already approved drugs that could be repurposed to treat other diseases. The system, he said, is rife with potential, particularly for autoimmune diseases. “… we describe a framework for integrating large-scale heterogeneous molecular and clinical data that can be used for other diseases, especially ones with no FDA-approved drug available for treatment,” the scientists concluded.
