Leukemia After Chemo/Radiation Not Directly Due to Therapy

Neil Osterweil

January 08, 2015

Cancer therapy may not be directly responsible for the development of poor-prognosis therapy-related acute myeloid leukemia (t-AML) or myelodysplastic syndrome (t-MDS), new research suggests.

Conventional clinical wisdom holds that chemotherapy and radiation induce genome-wide DNA damage that in a small percentage of patients causes malignant transformation of hematopoietic cells, leading to MDS or AML.

But evidence from a study of patients with t-AML and t-MDS suggests that rather than causing mutations, cytotoxic therapies may instead allow for the emergence and proliferation of existing treatment-resistant mutations in the tumor-suppressor gene TP53.

"What's happening is that P53 mutations are there before patients ever see chemotherapy, and those mutations get selected for," said investigator Daniel C. Link, MD, from the Genome Institute at Washington University in St. Louis, Missouri, in an interview with Medscape Medical News.

Dr Link and colleagues reported their findings in a recent online publication in Nature.

"This is a very important paper because it actually changes our view of how therapy-related leukemia occurs pathophysiologically," commented Richard M. Stone, MD, program director for adult leukemia at the Dana-Farber Cancer Institute and professor medicine at Harvard Medical School in Boston, Massachusetts, who was not involved in the study.

"When I was a fellow, and up until this type of research, the notion was that if you got this type of chemotherapy, the cumulative damage to bone marrow DNA could lead to mutations. This, coupled with other data, suggests that we as human beings accumulate mutations stochastically or by random chance over the course of a lifetime," he said in an interview with Medscape Medical News.

Why the Difference?

Therapy-related AML and MDS are typically diagnosed 5 to 6 years after a patient is exposed to cytotoxic therapies, but they have been reported to occur as early as 10 months and as late as 16 years after exposure. They are notoriously resistant to treatment, and prognosis is poor, with median survival of approximately 7 to 8 months after diagnosis, according to the National Cancer Institute.

"This started out as a science question," Dr Link said. "We knew that therapy-related AML or MDS was quite different, with a worse prognosis, different cytogenetics, and high incidence of P53 mutations."

"So when we went into this, the question was how does chemotherapy exposure shape that process?" Dr Link continued. "Because that's the only difference" between therapy-related and other forms of these diseases.

To delve further into this mystery, the investigators sequenced the genomes of 22 patients with therapy-related AML and compared them with sequences from patients with de novo AML or secondary AML that developed from MDS in patients who had not received chemotherapy other than hydroxyurea.

They found that the total number of somatic single-nucleotide variants and percentage of chemotherapy-related purine-pyrimidine substitutions (transversions) were similar between the two disease states. This finding indicates that chemotherapy and radiation do not induce DNA damage across the genome.

This led the researchers to consider whether something else was going on, and they hit on the idea that chemotherapy and radiation may allow for the emergence of P53 mutations that were already present in the cells of patients.

"This grew from work we published in Cell a few years ago: the idea that as you age, your hematopoietic stem cells are acquiring mutations. We did some mathematical modelling to determine that in a 50-year-old, about 50% of us would have least one of our 20,000 stem cells with a mutation in P53," Dr Link told Medscape Medical News.

To explore whether the pattern of genes that were frequently mutated in t-AML or t-MDS differed from that seen in de novo AML or MDS, the investigators looked for specific mutations in a panel of 149 genes related to AML and MDS in a group of 89 additional patients with t-AML or t-MDS.

The combined cohort (22 patients with whole genome sequencing plus the 89 additional patients) included 52 patients with t-AML and 59 with t-MDS. The investigators found that 55% of the samples had abnormalities in chromosomes 5 or 7, or complex cytogenetics.

They then compared the data with those from 199 previously reported genome or exome studies from patients with de novo AML and 150 patients with de novo MDS.

"As reported previously, TP53 mutations are significantly enriched in t-AML/t-MDS compared with de novo AML/MDS," the investigators write. "Interestingly, mutations of ABC transporter genes, a subset of which have been implicated in chemotherapy resistance, are also enriched in t-AML versus de novo AML. On the other hand, several well-defined driver gene mutations (that is, DNMT3A and NPM1) were significantly less common in t-AML. Thus, although the total mutation burden is similar, a distinct subset of mutated genes is present in t-AML/t-MDS."

Age-Related Changes

Combined with their previous observations about the accumulation of TP53 mutations with age, the authors hypothesized that hematopoietic stem/progenitor cells carrying age-related mutations are preferentially expanded or enriched after chemotherapy and/or radiation.

This hypothesis was supported by evidence that in four patients with t-AML, specific TP53 mutations found at diagnosis were also present in blood or bone marrow samples taken from the same patients 3 to 6 years earlier. In two of the four patients, the TP53mutations were identified before they had received any chemotherapy.

Although these findings do not have immediate clinical implications, they suggest that in the near future genomic scans could be used to predict risk for t-AML/t-MDS, Dr Link says.

"What we're looking at right now is whether we could use some of our sequencing techniques to determine the burden of stem cells carrying P53 or other leukemia-associated mutations in people who have seen a lot of chemotherapy, measure the size of their clone over time, and see if that can predict whether people are at risk," he said.

Dana-Farber's Dr Stone commented that "in retrospect this make sense, because if we take patients with P53 mutations who walk in the door and give them chemotherapy or even a bone marrow transplant, they do miserably, unfortunately, because it's a very treatment-resistant clone," he said.

Amir T Fathi, MD, a leukemia specialist at the Massachusetts General Hospital Cancer in Boston and assistant professor of medicine at Harvard Medical School, told Medscape Medical News that the study offers new insights into how t-AML and t-MDS may emerge but would not change his current clinical practice.

"If I saw a P53 mutation in a patient with a therapy-related AML, my approach would be to try a transplant, but even with transplants, those patients do worse, and their likelihood for long-term outcomes is poor. Nonetheless, you try to be more aggressive in patients who have these high-risk alterations," he said.

Dr Fathi was not involved in the study but agreed to comment on it for Medscape Medical News.

The study was supported by the National Institutes of Health and Leukemia & Lymphoma Society. Dr Link, Dr Stone, and Dr Fathi have disclosed no relevant financial relationships.

Nature. Published online December 8, 2014. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.