Is testicular irradiation necessary for patients with acute lymphoblastic leukemia and testicular relapse?

Improved cure rates for childhood cancer have created new opportunities, as well as challenges, for pediatric oncologists to develop treatments that could further enhance survival while reducing late sequelae. The testes have long been considered a pharmacological sanctuary site in the treatment of acute lymphoblastic leukemia (ALL), mainly because of the frequency of testicular relapse during or after chemotherapy. Hence, overt testicular leukemia at diagnosis or relapse was routinely treated with local irradiation. The addition of high-dose methotrexate (HDMTX) to systemic chemotherapy regimens since the 1970s has significantly reduced the risk of testicular relapse in patients with newly diagnosed ALL.^1,2 This result prompted us to omit testicular irradiation for patients with testicular leukemia at diagnosis who had received intensive chemotherapy, including HDMTX, in Total Therapy studies – if they lacked overt evidence of testicular disease after remission induction (biopsy was replaced by sonography after 1998).³ Of the 19 patients (2.3% of 811 consecutive boys with newly diagnosed ALL) who were treated by this approach, only 2 received irradiation for residual testicular disease; both subsequently died either from isolated central-nervous-system relapse or from combined hematologic and central-nervous-system relapse. Among the remaining 17 boys, only 1 developed a combined hematologic and testicular relapse, and there were no significant differences in event-free or overall survival rates between boys with or without testicular leukemia at diagnosis.³ This outcome has contributed to current widely accepted view that testicular leukemia at diagnosis is not an indication for testicular irradiation.

Late-onset isolated testicular relapse is another issue altogether. In a report from the Dutch Late Effects Study Group, all 5 boys with late testicular relapse, detected 8 to 71 months (median, 22 months) after completion of initial chemotherapy, remained alive in second remission for 1 to 15 years (median, 4 years) after retrieval chemotherapy including HDMTX (5 to 12 g/m²) without testicular irradiation.⁴ Encouraged by this experience, Barredo and colleagues⁵ at the Children’s Oncology Group designed a response-adapted treatment regimen for patients with late-occurring isolated testicular relapse. The primary objective was to establish the efficacy and toxicity of intensive systemic chemotherapy without testicular irradiation in this cohort of patients. Because of the rarity of testicular relapse and data from the Children’s Cancer Group trials for isolated testicular relapse (diagnosed between 1989 and 1995) showing a superior outcome for patients with an initial complete remission duration of 18 months or more (5-year survival, 61% vs. 43% in the comparison group with a shorter duration of initial remission), Barredo et al.⁵ selected a remission duration ≥ 18 months to define late testicular relapse (as opposed to the more conventional “off-therapy relapse”⁶).The treatment regimen included 9 courses of HDMTX (5 g/m² per course) and 14 courses of high-dose dexamethasone (10 mg/m² per day for 7 days) in combination with vincristine, as monthly pulse treatment with vincristine and corticosteroid had been shown to reduce the frequency of testicular relapse.⁷ In an effort to preserve testicular function, testicular irradiation was omitted for patients with normal testicular size or a negative biopsy for testicular leukemia at the end of a 4-week remission induction phase, and the cumulative cyclophosphamide dose (including doses given in the initial clinical trial for newly diagnosed leukemia) was capped at 6.4 g/m². Only patients with residual testicular leukemia, as evidenced by biopsy at the end of induction, received 24 Gy bilateral irradiation during consolidation therapy.

Of the 42 patients enrolled in this trial between 2004 and 2011, 2 had T-ALL, 1 of whom remained alive at 4.8 years. Analysis of the study was limited to 39 evaluable patients with B-ALL, 11 of whom received testicular irradiation after a positive biopsy; 28 were not irradiated owning to the resolution of testicular enlargement (n=14) or no morphologic evidence of leukemia in the biopsy sample (n=14). Major adverse events included 3 hematologic relapses among the 11 irradiated patients, and 6 isolated testicular relapses, 2 hematologic relapses, 1 death, and 1 second neoplasm among the 28 non-irradiated patients. There was no significant difference in 5-year event-free survival (72.7%±14.4% vs. 60.7±11.5%, P=0.595) or overall survival (72.7%±14.4% vs. 71.4%±10.6%, P=0.895) between the 11 irradiated and 28 non-irradiated patients.

Although eliminating testicular irradiation appeared successful in the above study (18 non-irradiated patients, or 46% of all enrolled patients, remained in prolonged second continuous complete remission), there was a trend toward improved 5-year event-free survival for the irradiated patients, despite their having residual testicular leukemia at the end of remission induction therapy. Importantly, 6 non-irradiated patients required retrieval therapy, which for some included allogeneic hematopoietic cell transplantation, for second testicular relapse. The small difference in event-free survival and overall survival rates among non-irradiated patients suggests that 2 or perhaps 3 of these children died after treatment for second testicular relapse. Whether elimination of testicular irradiation contributed directly to these presumed deaths can only be conjectural. As pointed out by the investigators, this study was not designed to compare treatment with or without testicular irradiation, making it difficult to assess the impact of this intervention on clinical outcome.

The Barredo study would have been more informative had it been possible to compare prognostic indicators between the irradiated and the non-irradiated cohorts. The presence of minimal residual disease (MRD) at diagnosis of testicular relapse was correlated with a poor outcome for the entire cohort: 2 relapses (both hematologic) occurred in 2 of the 3 patients with a positive MRD finding (0.01% and 0.1%) as compared to only 7 relapses (5 hematologic and 2 testicular) in the remaining 26 patients without detectable MRD.⁵ Although MRD was not predictive of subsequent testicular relapse, regardless of treatment with or without testicular radiation, it is unclear whether MRD results were balanced between the 2 cohorts of patients. The median and range of the initial remission durations were not reported or compared. Information on the unilateral versus bilateral testicular involvement at relapse, and on-therapy versus off-therapy relapse, also was either not available or not analyzed. Moreover, whether the presence of occult testicular leukemia after remission induction is predictive of subsequent testicular relapse in non-irradiated patients could not be determined because all 14 patients with resolution of testicular enlargement lacked a testicular biopsy after remission induction. Finally, this study could not address the need of irradiation for testicular relapse in patients with T-cell ALL or those with B-ALL in early relapse (less than 18 months after initial diagnosis).

I applaud the efforts of Barredo and colleagues to improve the quality of life of their patients. It remains to be seen if studies of irradiation for testicular relapse will be relevant in the context of CAR T-cell therapy, which appears able to eradicate extramedullary disease in at least some patients with relapsed ALL.⁸

Abbreviations

ALL

acute lymphoblastic leukemia

HDMTX

high-dose methotrexate

MRD

minimal residual disease