Genotype Predicts Treatment-Related Mortality in AML

Roxanne Nelson

December 11, 2012

ATLANTA, Georgia — A specific genetic marker might be associated with less toxicity from chemotherapy in some pediatric populations with acute myeloid leukemia (AML).

According to data presented here at the American Society of Hematology 54th Annual Meeting, the presence of single-nucleotide polymorphism (SNP) rs16754 in the WT1 gene correlates with improved outcomes in certain ethnic populations — predominantly Asian and black children with AML.

Dr. Phoenix Ho

The WT1 gene plays a role in the development of normal blood and genitourinary tissue, explained lead author Phoenix Ho, MD, attending physician at Seattle Children's Hospital and research associate at the Fred Hutchinson Cancer Research Center in Seattle, Washington. The abnormal overexpression of WT1 is present in 81% of pediatric AML patient samples, and mutations in the WT1 zinc-finger regions are present in 8% of pediatric AML patients.

Dr. Ho spoke at a press briefing held in advance of his presentation.

This study could have significant implications for treatment, said William G. Woods, MD, director of pediatric hematology/oncology and Daniel P. Amos Children's Chair of the Aflac Cancer Center and Blood Disorders Service at Children's Healthcare of Atlanta in Georgia, who moderated the press briefing.

It might be that patients with "the normal SNP can get by with fewer doses," he told Medscape Medical News.

Ethnic Variations

The researchers note that the successful treatment of childhood AML requires multiagent chemotherapy that starts with an intensive induction regimen. Treatment-related mortality remains a significant concern, but it can be affected by various factors specific to the patient.

They also note that the outcomes vary in adults and children with AML of different racial and ethnic groups. "This is probably due to both biologic and nonbiologic factors," explained Dr. Ho.

The frequency of SNP rs16754 varies significantly in the different populations.

Dr. Ho and colleagues found the SNP in 25% of white patients, 21% of black patients, 34% of Hispanic patients, and 53% of Asian-American patients. "These frequencies are what we would see in the normal populations," he noted.

Less Treatment-Related Mortality in Certain Groups

The researchers assessed the relation between the rs16754 genotype and 5-year rates of overall survival, complete remission, relapse risk, and treatment-related mortality. To examine the SNP rs16754 genotype, they directly sequenced WT1 exon 7 on 492 available diagnostic specimens from AML patients enrolled in the Children's Cancer Group CCG-2961 trial.

"It turns out that our SNP is not significantly associated with improved remission rates or relapse rates.... But when we looked at treatment-related mortality and stratified it by ethnicity, it turned out that children who were positive for the SNP and of Asian or African ancestry had 0 toxic deaths," he pointed out.

This is in contrast to children of those ethnic groups who were negative for the SNP; they had high rates of toxic deaths, Dr. Ho added.

"Since these patients are treated with multiagent chemotherapeutic agents, it can be difficult to tease out which agents are most responsible for the toxicity," he explained.

Patients who had at least 1 minor (G) allele were considered positive for the SNP, whereas patients homozygous for the major (A) allele were considered negative for the SNP. The researchers stratified SNP genotype by ethnicity and analyzed the outcomes.

All patients in the trial received initial induction therapy with an intensively timed hybrid regimen of idarubicin/daunorubicin, dexamethasone, cytarabine, thioguanine, and etoposide. Patients who had at least a partial remission (<30% marrow blasts) were then randomized to a repeat of that regimen or to idarubicin, fludarabine, and cytarabine.

Because most treatment-related mortality occurs during the first 2 cycles of chemotherapy, the researchers evaluated it separately in the 2 study groups.

Overall survival at 5 years was better in the 28% of patients who were positive for the SNP than in those who were negative (61% vs 44%; P = .009). However, other outcome measures, such as disease-free survival and response rates, did not appear to be significantly affected by SNP status.

In SNP-positive ethnic subgroups, 5-year overall survival was better in Hispanic patients (49% vs 37%; P = .433), black patients (56% vs 30%; P = .207), Asian patients (88% vs 29%; P = .037), and white patients (65% vs 48%; P = .051). Statistical significance was approached in the white patients and was reached in the Asian patients.

For the entire study population, treatment-related mortality rates did not vary by rs16754 genotype. However, the combined treatment-related mortality rates of black and Asian patients were significantly lower in SNP-positive than in SNP-negative patients. Among the black population, SNP-negative patients had significantly higher rates of treatment-related toxic death than SNP-positive patients (25% vs 0%). The same was true in the Asian population (43% vs 0%).

These results suggest that the protective effect of SNP rs16754 in decreasing chemotherapy-related toxicity in pediatric AML patients is more pronounced in those of African and Asian descent, the researchers note.

The researchers would like to validate the ethnic-specific association between SNP status and treatment-related mortality in a large contemporary clinical trial. They would also like to evaluate the relation between SNP rs16754 and treatment response for other cancer types. explained Dr. Ho.

American Society of Hematology (ASH) 54th Annual Meeting: Abstract 1385. Presented December 10, 2012.