Researchers at St. Vincent’s Comprehensive Cancer Center have identified a new gene signature that may help doctors determine which pre-leukemia patients will benefit from treatment with Celgene's myelodysplastic syndrome drug Revlimid (lenalidomide).
Although the findings need to be replicated, researchers hope eventually to develop a “simple diagnostic” for this indication.
The researchers identified 30 red cell-related genes that could help personalize treatment with Revlimid for pre-leukemia patients. The study, which was published in the February issue of the journal PLoS Medicine, aimed to develop a method for predicting Revlimid response “in order to avoid unnecessary toxicity in patients unlikely to benefit from treatment.”
Although patients with an interstitial deletion of chromosome 5q have a high rate of response to Revlimid, most MDS patients lack this deletion. However, a study published in the January edition of Blood, showed that 26 percent of patients without the 5q deletion did not need blood transfusions after treatment with the drug.
From this study, “we postulated those patients must share some other genetic profile, which led us to the discovery of the 30 under-expressed red cell-related genes.” Naomi Galili, director of St. Vincent’s Bone Marrow Translational Laboratory, said in a statement. Researchers at St. Vincent's Comprehensive Cancer Center collaborated on this study with scientists at the Broad Institute of MIT and the Dana Farber Cancer Institute.
Ultimately, the researchers are hoping that their findings will lead to an FDA-approved test to help guide Revlimid treatment.
“Once we can confirm the 30-gene signature in a larger group of patients, we hope to develop a simple diagnostic test,” Azra Raza, director or the MDS program at St. Vincent's Comprehensive Cancer Center, told Pharmacogenomics Reporter in an e-mail. According to Raza, the confirmatory trial design has been approved and Celgene is willing to provide the drug and “partial support” for the study.
When it finishes the confirmatory studies, the St. Vincent’s team “hope[s] to take the test through FDA approval,” Raza said, adding that a diagnostic kit would be developed by Celgene since it owns the intellectual property for the drug.
A Celgene spokesperson noted, however, that while the St. Vincent's findings are interesting, they are still very early stage. “There are no current plans for a diagnostic in MDS,” Greg Geissman, associate director of public relations at Celgene, told Pharmacogenomics Reporter last week.
In their PLoS Medicine paper, the researchers used Affymetrix's GeneChip technology to compare the expression profiles of patients who responded to Revlimid with those who did not, regardless of their 5q status. The researchers identified a cohesive set of erythroid-specific genes from a set of 16 pretreatment bone marrow aspirates from MDS patients without 5q deletions. The researchers validated the results in an independent set of 26 samples.
Since the genes identified decreased expression in responders, this suggested that a defect in erythroid differentiation underlies Revlimid response. “Consistent with this observation, treatment with [Revlimid] promoted erythroid differentiation of primary hematopoietic progenitor cells grown in vitro,” the study authors wrote.
“These studies indicate that lenalidomide-responsive patients have a defect in erythroid differentiation, and suggest a strategy for a clinical test to predict patients most likely to respond to the drug.”
According to Raza, the clinical challenge was to find a method for recognizing the 25 percent of potential responders without the 5q deletions before starting therapy in order to prevent the 75 percent of potential non-responders from having to experience the adverse effects of a drug that will not help them.
Using Affy's platform to compare the gene expression profiles of responders and non-responders to Revlimid, researchers identified 50 genes that were regularly under-expressed by all the responders, regardless of their 5q status.
“Most of the under-expressed genes turned out to be those involved in erythropoiesis or red cell formation,” Raza said. “Note that Revlimid response is predominantly erythroid in that responders improve their anemia and patients stop requiring transfusions. Thus, biology appeared to recapitulate the clinical observation.”
Although this finding meant that researchers could now more accurately identify potential responders to Revlimid before starting therapy, no diagnostic exists that could help doctors make prescribing decisions in a prompt and clinically useful fashion.
“So we developed a more user-friendly version of the test by identifying the top 30 genes out of the 50 [that] best distinguish between the responders and non-responders using a quantitative PCR assay available commercially using Luminex beads,” Raza said.
According to Raza, Luminex bead hybridization is a simpler assay to measure gene expression than microarrays, and the results of the test are “available within hours of procuring the samples.”
The researchers validated the accuracy of the Luminex-based tests in a new Revlimid-treated cohort in which the outcome of treatment was unknown. The study revealed that a test based on Luminex's platform accurately predicted which MDS patients without 5q deletions would benefit from Revlimid treatment approximately 90 percent of the time.
“These studies indicate that [Revlimid]-responsive patients have a defect in erythroid differentiation, and suggest a strategy for a clinical test to predict patients most likely to respond to the drug,” the study authors concluded in the paper. “The experiments further suggest that the efficacy of Revlimid, whose mechanism of action in MDS is unknown, may be due to its ability to induce erythroid differentiation.”
According to Raza, these new findings are particularly timely since the incidence of MDS is on the rise in the US. While there were only 143 cases of MDS in 1973, Raza places current MDS incidence rates between 10,000 and 20,000 per year.
The incidence “is expected to increase in the next decade if this trend continues,” he added. “This increase in number is multi-factorial; aging of the population and precise diagnosis being two major contributing causes. MDS is now more common than de novo acute myeloid leukemia in the older age groups.”
According to the Leukemia & Lymphoma Society, the estimated incidence of MDS in the US is unknown. However, the group estimates that the number of new MDS cases annually is similar to the incidence of AML. In 2006, there were approximately 12,000 new AML cases in the country.
While people at any age may be afflicted with MDS, the highest incidence of disease manifests in men over 65 years of age. The causes for developing MDS are unknown, although studies have shown those who have worked with benzene, had exposure to radiation or chemotherapy treatments, or have certain genetic conditions such as Down syndrome may be at high risk.
Revlimid, approved by the FDA in December 2005, is related to thalidomide, and is similarly teratogenic. Therefore, Revlimid cannot be prescribed to pregnant women due the risk of causing defects or malformation in unborn children. Other side effects with Revlimid include thrombosis, pulmonary embolus, hepatoxicity, and bone marrow toxicity resulting in neutropenia and thrombocytopenia. A major dose-related toxicity from Revlimid is myelosuppression.
In addition to Revlimid, there are two other FDA-approved drugs for MDS: Pharmion's Vidaza and SuperGen's Dacogen.
“The principal challenges now are to match the right drug to the individual patient,” Raza said. “[Our] discovery will help tailor therapy to suit each MDS patient individually who presents with anemia without the chromosome 5 abnormality.”