NEW YORK – Researchers and clinicians at Memorial Sloan Kettering are hoping to start generating answers to a set of crucial questions about the molecular development of cancer through an effort called the clonal hematopoiesis clinic.
The program, now in its third year, has been collecting genomic sequencing and other longitudinal data and tracking the outcomes of patients who have cancer-associated mutations present in their blood or bone marrow but don't yet show any signs of a hematologic malignancy like leukemia.
The clinic is part of a larger initiative called Precision Interception and Prevention in which MSK aims to employ existing tools and develop new strategies to catch cancer early or prevent it completely.
Clonal hematopoiesis (CH), which describes the clonal emergence of some of the same mutations that are found in blood cancers but without a transformation to malignant disease, is not a new discovery. Evidence has accumulated over the last decade that CH is both very common and is linked to an increased risk of certain blood cancers, especially myelodysplastic syndrome and acute myeloid leukemia, as well as cardiovascular disease, heart attacks, and strokes. However, it doesn't lead to these outcomes in all, or even most of those who develop it.
"There was this longstanding notion … that you could have clonal blood cells without disease," said Ross Levine, the CH clinic's director and one of the first researchers to discover links between CH and blood cancer.
Around 2012, he noted, "it became clear that recurrent mutations in known leukemia genes can occur in people without overt malignancies, and that if you follow these people long enough, a subset of them develop hematologic cancers." In subsequent studies of larger cohorts "the hypothesis emerged that maybe these people represented an at-risk population kind of like germline cancer-risk mutation carriers," Levine added.
Only recently have programs tried to discover mechanisms that explain exactly how CH mutations elevate the risk of developing cancer (or non-cancer diseases), why they portend malignancy in some individuals but not others, and whether treatments can be adapted or developed to halt this process.
Levine said that the MSK team believes its effort was the first of its kind when it launched in 2018. Numerous similar projects have now followed including efforts at MD Anderson, Dana Farber, University of North Carolina, University of Chicago, and Weill Cornell Medicine. "This whole field is actually quite collaborative, and there's a lot of note comparing between the different efforts, which has been gratifying," Levine said.
At MSK, the creation of a clinical CH tracking program was driven by accumulating data from the center's solid tumor next-generation sequencing efforts.
"Because of our MSK-IMPACT program, where we do lots of tumor somatic profiling and we do blood as the germline comparator, we were able to publish a paper … in 2017 in Cell Stem Cell where we could not only show that [CH] occurs at a high frequency — 25 percent of everyone undergoing MSK-IMPACT — but that it also had adverse outcome. And that really locally galvanized our group to start thinking about what a clinic would be like," Levine said.
Unlike some other CH tracking programs, MSK's effort is clinically integrated and driven. Participants are recruited from the pool of solid tumor patient who receive MSK-IMPACT as part of their medical care at the center. Those whose germline data reveals the presence of CH are contacted by their physician to offer them entry, and almost 150 have been brought in so far.
"What we do is we notify clinicians of the patients that we think are at the highest risk," Levine said. "And obviously if someone's got a short life expectancy or metastatic cancer, we're not going to ask someone who is on fifth-line therapy. We're looking for people we think have a long life expectancy from their current cancer, where … a second malignancy is a relevant concern to discuss with them."
Prior to the launch of the clinic, these findings would have been considered incidental, but with the program in place, patients have the option to enroll in the clinic where they have their blood tested on a regular basis using a hematologic-specific version of the MSK sequencing test. "Currently, we're largely doing sampling every six months, both for research, if people are willing, as well as to follow them clinically," Levine explained.
"I think we're all still trying to figure out what these clinics are going to look like," he added. "The idea of seeing people at risk and finding ways to counseling them, it's like BRCA before you had drugs. And our view was always that this was sort of a phase zero."
In the nearer term, the CH clinic offers the potential for earlier detection and better treatment for those that do develop a malignancy.
According to Levine, the group has already learned a lot in this vein. For example, "we are seeing a lot more people who when you find they have CH and when you actually see them, some of them already have disease and just no one noticed," he said.
Another potentially positive near-term impact is that individuals can be better counseled regarding cardiovascular risk, Levine added. "We're not saying that you would give people drugs to reduce their cholesterol if they wouldn't otherwise need them. But it's a good reminder to assess their cardiovascular risk more holistically."
In the long term, tracking how CH mutational profiles change over time and comparing individuals who develop a leukemia or associated syndromes to those who don't could offer a framework to better personalize care in the future by identifying which individuals actually need to receive close monitoring or, ideally, preventative treatment.
Levine said the team has another paper coming out soon that will refine the differentiation of riskier and less-risky CH findings, so that that clinicians can narrow down a subset of individuals who need truly need close follow-up or other intervention.
With that in place, the team can start thinking about exploring therapeutics in this space — launching clinical trials of treatments that could block this progression. But it will be important to recognize, Levine said, that potentially "treating" CH is not like treating cancer. "We're going to need cleaner drugs with wider therapeutic indices and the dose and the mechanism will be different. But that's not going to stop us from trying," he added.
Levine also argued that taking another shot at developing preventive strategies is going to have to be much more genomically integrated than it has been in the past. " We can't say, 'Oh it's an inflammatory state, let's try an anti-inflammatory on everybody.' I think we're going to need genotype-selective approaches for the most common mutations or for ones that are directly targetable. And we're going to have to do this based on rigorous preclinical science."
Levine cited some early hypotheses that could represent the first new shots on goal that the group takes, including targeting IDH and JAK2 mutations, which already have therapeutics being advanced to target them.
"We've had many conversations to try to think about how to do those trials … And I think there is a lot of interest and we've had some interest from pharmaceutical companies," he said.
Unfortunately, these alterations in IDH and JAK are seen in only a small proportion of CH cases: about three or four percent of all clonal hematopoiesis. "It would be nice to do those trials where we have drugs that are approved, and we know how to use them, and where I think the therapeutic index is favorable. But those are not going to be easy to recruit, nor really relevant to the majority of people with CH," Levine said.
In terms of other future plans for the clinic, Levine said the team is grappling with whether and how its approach might be broadened. "We think we can actually counsel people not just locally, but nationally who have been found to have sequencing. I think on a broader population level, the question will be, outside of cancer, how many people are being sequenced? Should we sequence them? We just don't know yet. And we hope that clinics like this will help learn the rules before we start thinking about this being relevant to the general population or not."
Finally, he said, the team is hoping to be able to develop and implement ultrasensitive genomic assays that can uncover information the current hematologic version of the center's core assay cannot.
"We want to not only identify what lesions people have but monitor them for evidence of clonal evolution [because] it's not the CH mutations that really denote all of their risk necessarily, it's that they acquire additional events," Levine said.
"We're now assessing the feasibility of that, and ideally how it could be exported more broadly," he added.