Predictive Factor Analysis of Response-Adapted Radiation Therapy for Chemotherapy-Sensitive Pediatric Hodgkin Lymphoma: Analysis of the Children's Oncology Group AHOD 0031 Trial

Purpose

To evaluate whether clinical risk factors could further distinguish children with intermediate-risk Hodgkin lymphoma (HL) with rapid early and complete anatomic response (RER/CR) who benefit significantly from involved-field RT (IFRT) from those who do not, and thereby aid refinement of treatment selection.

Methods and Materials

Children with intermediate-risk HL treated on the Children's Oncology Group AHOD 0031 trial who achieved RER/CR with 4 cycles of chemotherapy, and who were randomized to 21-Gy IFRT or no additional therapy (n=716) were the subject of this study. Recursive partitioning analysis was used to identify factors associated with clinically and statistically significant improvement in event-free survival (EFS) after randomization to IFRT. Bootstrap sampling was used to evaluate the robustness of the findings.

Result

Although most RER/CR patients did not benefit significantly from IFRT, those with a combination of anemia and bulky limited-stage disease (n=190) had significantly better 4-year EFS with the addition of IFRT (89.3% vs 77.9% without IFRT; P=.019); this benefit was consistently reproduced in bootstrap analyses and after adjusting for other prognostic factors.

Conclusion

Although most patients achieving RER/CR had favorable outcomes with 4 cycles of chemotherapy alone, those children with initial bulky stage I/II disease and anemia had significantly better EFS with the addition of IFRT as part of combined-modality therapy. Further work evaluating the interaction of clinical and biologic factors and imaging response is needed to further optimize and refine treatment selection.

Introduction

For children with Hodgkin lymphoma (HL), the combination of chemotherapy and involved-field radiation therapy (IFRT) produces excellent cure rates, with event-free survival (EFS) exceeding 85% in most recently published trials 1, 2, 3. However, because of therapy-related late toxicity 4, 5, 6, selection of treatment intensity remains a clinical challenge, especially with respect to optimally refining the use of radiation therapy (RT). It would be ideal to find a group of patients for whom RT can be omitted without a significant increase in the rate of relapse, and conversely, to provide RT only to patients for whom there is a clinically significant benefit. Utilization of RT based on end of chemotherapy anatomic response alone can significantly increase the risk of relapse (7), and consequently recent clinical trials have evaluated early chemotherapy response assessment as a means of identifying patients with highly chemo-sensitive disease for whom RT can be omitted without a clinically significant loss of disease control.

The Children's Oncology Group (COG) AHOD 0031 trial evaluated a response-based treatment approach (2). Among intermediate-risk patients with both rapid early anatomic response and complete response (RER/CR; determined by CT imaging), IFRT did not produce a significant improvement in 4-year EFS compared with adriamycin, bleomycin, vincristine, etoposide, prednisone, and cyclophosphamide (ABVE-PC) chemotherapy alone (87.9% vs 84.3%, P=.11) (2).

However, this trial enrolled a clinically heterogeneous group of patients (ie, IA with bulk-IVA) with a broader range of risk factors than most adult “intermediate risk” trials. Consequently, there is some uncertainty regarding whether 4 cycles of chemotherapy alone was adequate for all RER/CR patients and whether treatment could be more optimally refined according to individuals' clinical characteristics.

The present analysis was undertaken to evaluate whether clinical risk factors could further distinguish intermediate-risk RER/CR patients who seemed to benefit significantly from IFRT from those who did not, and thereby aid refinement of treatment selection.