Response:Solid cancers after allogeneic hematopoietic cell transplantation

Response:

We appreciate the comments from Dr Hsieh et al, who raise an important issue regarding our recent publication.¹  They point out the apparent lack of a dose-response relationship between total body radiation dose and second cancer risk, where fractionated doses of less than 12 Gy or a single dose of 10 Gy were equally associated with elevated risks of solid tumors. This finding contrasts with data from the atomic bomb survivors in whom a near-linear dose response was demonstrated, albeit for a much lower range. Although Hsieh et al point out that the HCT community is increasingly using lower doses of radiation in the preparative regimen (reduced intensity and nonmyeloablative regimens), several important differences between our population and the atomic bomb survivors must be kept in mind. Approximately 99% of exposed members of the Life Span Study cohort of atomic bomb survivors received radiation doses less than 2 Gy,²  which is considerably lower than the doses administered to transplant recipients in our study receiving myeloablative conditioning regimens. Other differences include the immunocompromised status of HCT patients, and differences in type and quality of radiation received.

Our study included patients whose procedures were performed before 1996. During that period of time, most patients received myeloablative preparative regimens with very few receiving TBI doses less than 5 Gy. In the single dose TBI group, there were actually no second cancers observed among the only 205 patients whose TBI dose was less than 5 Gy. Among the 1729 patients who received single doses of radiation between 5 and 9.9 Gy, there were 10 second malignancies. However, among the 111 patients who received fractionated TBI at doses less than 8 Gy, 2 second malignancies occurred, among the 522 who received between 8 Gy and 9.9 Gy, 3 cases occurred, and among the 2077 who received between 10 and 11.9 Gy, 12 cases occurred. As indicated, very small numbers of patients and extremely small numbers of “expected” events in the lowest radiation groups (< 5 Gy single or < 8 Gy fractionated total dose) make meaningful analyses difficult. This is unfortunate, as one theoretical advantage of the low doses of TBI in current reduced-intensity preparative regimens is a lower risk of therapy-related cancers. We agree with Hsieh and colleagues that a study that could determine whether this is, in fact, the case would be important. Meaningful analyses of these lower dose TBI regimens should be possible once a larger number of patients transplanted since the late 1990s has been followed for 5 years or longer. We intend to address the issue at that time.

The incidence of second cancers presented in this article included patients who received HCT for nonmalignant disorders, such as severe aplastic anemia (SAA; 10% of patients) and hemoglobinopathies (3%). Our Poisson regression analyses were stratified by age, disease, and time since HCT, because risks differ based on these factors. As our report discussed,^1p1180 the risk of second cancers in patients with SAA was not significantly different from those with acute leukemia, and we found no interaction between conditioning radiation and primary disease with regard to risk of second cancer. Unfortunately, additional subset analyses were not possible because of the relatively small number of HCT patients who subsequently developed a second cancer.