Cancer Treatment May Elevate Risk of Acute Lymphoblastic Leukemia

Myeloid malignancies, such as acute myeloid leukemia (AML), are a well-established treatment-related complication of radiotherapy and chemotherapy for first cancers. But little has been known about the risk of acute lymphoblastic leukemia (ALL) following first cancer treatment.

Prompted by an influx of patients who developed ALL as a second cancer and case reports, the Department of Hematology and Medical Oncology at Cleveland Clinic Cancer Center conducted the first large-scale study of the risk of ALL in patients treated for first cancers.

Analyzing SEER data

Since second cancers occur only in a limited number of patients with first cancers, the study was purposefully designed to interrogate a large number of cancer patients followed over many years. The authors used the U.S. Surveillance Epidemiology and End Results (SEER) registry, the largest repository of cancer patient data in the world, which includes 10 million cases from 1973 to 2015.

The investigators analyzed the registry data using SEERaBomb, a free R open source program, developed by study co-investigator Tomas Radivoyevitch, PhD, specifically for estimating second cancer risks. First cancer cases of lymphoid lineage were excluded.

A total of 4,581,222 eligible first cancer patients were identified and classified according to treatment received: 821,004 (17%) received radiotherapy (RT) only; 571,035 (12%) received chemotherapy (CT) only; 488,530 (10%) received RT + CT; and 2,970,253 (61%) received neither.

In total, 849 patients developed ALL: 176 (21%) in the RT only group, 137 (16%) in the CT only group, 106 (12%) in the RT + CT group and 430 (51%) in the no RT/CT group. The risk analysis was based on the standardized incidence ratio (SIR) calculated as the ratio of the observed number of ALL cases (in the study population which is patients with first cancer) to the expected number of ALL cases (computed using age-specific rates from the general population, weighted according to the age structure of the study population).

Relative risk of ALL based on treatment type

Patients treated with RT or CT had an elevated risk of developing ALL in the first 10 years after first cancer diagnosis: the SIR for RT only was 1.49 (95% CI, 1.27-1.72), for CT only 3.10 (95% CI, 2.59-3.67) and for RT + CT 3.10 (95% CI, 2.53-3.75). No elevated risk was found in patients who didn’t receive cytotoxic treatment compared with the general population. “The greatest contribution to the risk of ALL appears to be in patients treated with CT or combined CT and RT with the risk contributed by RT exposure marginal in the latter group,” says the study lead investigator Sudipto Mukherjee, MD, PhD, MPH, Department of Hematology and Medical Oncology, Cleveland Clinic Cancer Center.

For comparison, the study analyzed data from breast cancer patients, a group with different modalities of treatment exposure. The SIR for developing ALL in the first 10 years were: RT + CT 3.49 (95% CI, 2.60-4.59), RT only 1.50 (95% CI, 1.08-2.03) and CT only 1.75 (95% CI, 1.08-2.03). Again, no elevated risk was found in patients who didn’t receive cytotoxic treatment compared to the general population.

In multivariate regression analyses of the entire cohort of cancer patients, younger age at diagnosis, male sex, year of diagnosis, and treatment modality (CT alone or CT+RT or RT alone versus no therapy) were significant predictors of ALL. “The risk of ALL occurrence after first cancers changes with follow-up time and seems to peak between the first and second year following treatment. After 10 years, the risk becomes comparable to the general population,” says Dr. Mukherjee.

Patients with hematologic first cancers (myeloid lineage or plasma cell dyscrasias) had the greatest risk of developing ALL as a second cancer. The SIR for AML patients treated with CT was 11.48, for multiple myeloma patients treated with CT, 7.44, and multiple myeloma patients treated with RT + CT, 14.41.

“There is a staggering increase in the risk of ALL in multiple myeloma and acute myeloid leukemia patients. We know that chemotherapeutic agents used in multiple myeloma and acute myeloid leukemia are leukemogenic but there are several other factors that could be contributing to ALL risk and need further in-depth investigation,” says Dr. Mukherjee.

Clinical implications

While the findings show a statistically significant association, they do not indicate causality, and additional studies are needed to validate them. “The findings from this study can be helpful in discussing treatment risks and benefits in daily clinical practice. With improvements in cancer therapies, there is a growing population of cancer survivors in the U.S., and studies such as this can hopefully educate cancer survivors and physicians alike about the subsequent risks of ALL,” Dr. Mukherjee says.

Future study

The researchers plan to delve more deeply into this area in subsequent studies, including investigating the risks of different chemotherapy regimens and whether a biologic mechanism plays a role in ALL development.

The findings were presented at the 2018 annual meeting of the American Society of Hematology.