Chronic Myeloid Leukemia as Secondary Malignancy Following the Treatment of Hodgkin Lymphoma: A Case Series

Discussion

The development of secondary CML has been reported in the setting of HL, NHL, and other lymphoid malignancies but it remains an uncommon event. Further, given the cytogenic and demographic similarity between de novo CML and secondary CML (3), it is difficult to discern if the above-reported cases are actually related to treatment of HL. In determining a relationship, several considerations must be made. These include the primary malignancy treated, the time between treatment inception and detection of CML, and possible mechanisms for the development of a secondary malignancy.

Regarding primary malignancy, all of the above cases of CML followed treatment of HL. One comprehensive review of secondary CML demonstrates that it is, in fact, more commonly attributed to treatment of HL than any other single malignancy (>25%). Of note, however, no individual study in this set demonstrated a statistically significant excess of CML compared to control populations (3). No mechanistic explanation for this was proffered, though it was suggested that this represents both a reporting bias (the seeming uncommonness of this event leading to over-reporting) as well as a detection bias (patients already being followed by a hematologist/oncologist are more likely to have other malignancies detected).

That said, there does appear to be a fairly consistent temporal relationship between treatment of HL and the development of CML. In the aforementioned review, that average difference was approximately four years. In our three cases, CML was detected within three years of initial treatment for HL. While it remains difficult to determine the precise chance of developing two unrelated primary malignancies within such a short time, the apparent consistency and short interval do suggest a relationship between treatment inception and development of CML.

As for the possible mechanism of secondary CML development, there does not appear to be an obvious increase in cases of CML after treatment with chemotherapy alone. Compared to AML and MDS for which exposure to DNA-damaging medications comprise between 10-20% of cases, rates of all secondary CML appear to increase only in regimens involving radiotherapy (4). Previous studies have demonstrated a 0.7% chance of developing CML after ABVD therapy alone as opposed to a 9.5% chance after combined ABVD and radiotherapy (5, 6). This may be attributed to the relatively increased amount of time that the progenitor stem cells of CML spend in a quiescent phase, especially compared to the fast-replicating progenitor cells of AML and MDS (3, 4). Whereas intercalating regimens are most effective at addressing actively dividing cells, radiation has been shown to produce double-strand breaks even in cells in the G₀ and G₁ phases (4). Interestingly, only one patient in our cases received ionizing radiation therapy though all did receive at least partial ABVD treatment.

Given the limited number of cases presented here, as well as the relative scarcity of reported secondary CML in the literature, it is not possible to determine whether the above represent true secondary malignancies. It is, however, likely that the prevalence of such disease is higher than previously reported due to underdetection. Given the increasing rates and duration of disease-free survival with first-line HL treatment, this case series serves to highlight the necessity of close surveillance of post-treatment HL patients for the development of treatment-related malignancies.

Interestingly, the third case represents by itself a very rare form of CML with atypical variant. While the hallmark of CML is a BCR-ABL translocation, in this instance the patient possessed a BCR-JAK2 translocation t(9;22) (p24;q11.2). Literature on such transformations remains sparse, however this specific translocation has been previously associated with both CML- and AML-like diseases (7-9). Novel fusions are being increasingly implicated in leukemias, including other JAK2 variants such as ETV6-JAK2 fusions. BCR-JAK2 fusions, however, appear to be rare events, possibly due to lack of routine testing for these mutations. In one reported case a BCR-JAK2 disease was not responsive to imatinib, likely due to lack of an active binding site for imatinib on an ABL kinase region (10). This serves to demonstrate that this fusion product may drive leukemic disease on its own and present a possible target for future therapies.