Declining Use of Radiotherapy in Stage I and II Hodgkin’s Disease and Its Effect on Survival and Secondary Malignancies

IJROBP 2012 International Journal of Radiation Oncology, Biology, Physics
Volume 82, Issue 2 , Pages 619-625, 1 February 2012

Concerns regarding long-term toxicities have led some to withhold radiotherapy (RT) for the treatment of Stage I and II Hodgkin’s disease (HD). The present study was undertaken to assess the use of RT for HD and its effect on overall survival and the development of secondary malignancies.

Methods and Materials

The present study included data from the Surveillance, Epidemiology, and End Results database from patients aged ≥20 years who had been diagnosed with Stage I or II HD between 1988 and 2006. Overall survival was estimated using the Kaplan-Meier method, and the Cox multivariate regression model was used to analyze trends.

Results

A total of 12,247 patients were selected, and 51.5% had received RT. The median follow-up for the present cohort was 4.9 years, with 21% of the cohort having >10 years of follow-up. Between 1988 and 1991, 62.9% had undergone RT, but between 2004 and 2006, only 43.7% had undergone RT (p < .001).

 The 5-year overall survival rate was 76% for patients who had not received RT and 87% for those who had (p < .001). The hazard ratio adjusted for other variables in the regression model showed that patients who had not undergone RT (hazard ratio, 1.72) was associated with significantly worse survival compared with patients who had received RT.

 The actuarial rate of developing a second malignancy was 14.6% vs. 15.0% at 15 years for those who had and had not undergone RT, respectively (p = .089).

Conclusions

The present study is one of the largest studies to examine the role of RT for Stage I and II HD. Our results revealed a survival benefit with the addition of RT with no increase in the development of secondary malignancies compared with patients who had not received RT. Furthermore, the present nationwide study revealed a >20% absolute decrease in the use of RT from 1988 to 2006

Introduction 

The treatment of early-stage Hodgkin’s disease (HD) has been a success story in oncology, when an incurable disease became curable in the 1960s with the use of external beam radiotherapy (RT). Since then, chemotherapy has similarly demonstrated significant improvements in the outcomes of patients with advanced disease, as well as those with early stages. The results have been so impressive that cooperative groups changed the treatment paradigm from improving survival to maintaining the same survival and reducing the morbidity by seeking to minimize the use of RT and chemotherapy. In the early 1990s, several publications revealed significant long-term complications associated with a combined modality approach involving full-dose chemotherapy and extended-field RT.

This led to the investigation of two treatment strategies. First, the use of combined modality therapy but with a reduction in the irradiated volume, radiation dose, number of chemotherapy cycles, and number of chemotherapy agents; and second, the use of chemotherapy alone with additional cycles and the elimination of RT from the treatment paradigm. The proponents of the chemotherapy-alone strategies believed improved overall survival (OS) would be seen owing to the reduction in late radiation-related mortality. Furthermore, any increase in relapses seen with chemotherapy-alone strategies would not affect survival, because these patients could be salvaged with additional chemotherapy and stem cell transplantation/

These two approaches were then tested in several prospective Phase III trials. All 5 trials showed a significant improvement in disease control with the addition of RT; however, this did not translate into a benefit in OS for those with Stage I and II HD. These trials were limited by the low patient numbers and limited follow-up that might have precluded them from demonstrating any significant survival benefit or any detriment from delayed radiation morbidity.

We undertook the present study to determine in a large population-based cohort whether RT would be associated with a survival benefit for those with Stage I and II HD. Furthermore, we examined the overall trends in the use of RT in early-stage HD and the effect of RT on the second malignancy rate compared with patients who had not undergone RT.

The present population-based study examined whether RT affected the survival outcomes of patients with HD. We found a statistically significant survival benefit for patients with Stage I and II HD. Furthermore, among the cohort who had not undergone RT, no increase was found in the development of secondary malignancies compared with the cohort without RT.

Several modern randomized studies have examined chemotherapy vs. chemotherapy plus RT for HD. The European Organization for the Research and Treatment of Cancer H9-F study was a three-arm study in which patients with favorable-stage disease who were complete responders to 6 cycles of epirubicin, bleomycin, vinblastine, and prednisone were randomized to 36 Gy of involved-field RT, 20 Gy of involved field RT, or no RT. An interim analysis revealed a 4-year event-free survival rate of 87% in the 36-Gy arm, 84% in the 20-Gy arm, and 70% in the no-RT arm (p < .001). The chemotherapy-alone arm was subsequently closed owing to the greater than expected number of relapses that met the proposed early stopping rule.

The Grupo Argentino de Tratamiento de la Leucemia Aguda performed another trial in which patients with Stage I and II HD were treated with six cycles of cyclophosphamide, vinblastine, procarbazine, and prednisone and then randomized to receive 30 Gy of involved-field RT vs. no RT. With a median follow-up of 84 months, the disease-free survival rate was significantly different at 71% vs. 62% for patients with vs. without RT.  A Children’s Cancer Group study also examined patients with HD who were complete responders to chemotherapy and randomized them to receive involved RT or no RT. An interim analysis revealed an unacceptable number of failures in the patients who had not undergone RT, and the trial was stopped early.

Our study, with >12,000 patients, is one of the largest cohorts of patients with HD. Our results revealed that RT led to improved OS and CSS for both those with Stage I and II HD. This is the first study to demonstrate a survival benefit for those with Stage I and II HD with the addition of RT. We hypothesized that the known improvement in local control from the addition of RT translated into an improvement in survival.

 A Cochrane Review examined 9,312 patients from 37 trials and revealed an improvement in the chemoradiotherapy arm compared with the chemotherapy-alone arm (odds ratio, 0.62; 95% confidence interval, 0.44–0.88; p = .006). Furthermore, the second malignancy rate was not significantly increased in early stage patients for those who had undergone chemoradiotherapy vs. chemotherapy alone. Another meta-analysis included 1,245 patients from five randomized trials and revealed an improved hazard ratio of 0.40 (95% confidence interval, 0.27–0.59) for patients receiving chemoradiotherapy compared with chemotherapy alone. Our results are in accordance with the findings from these meta-analyses of early-stage patients.

Concerns regarding long-term toxicity with RT have led some investigators to favor chemotherapy-alone strategies for those with favorable-risk HD. However, as mentioned, all clinical trials investigating this approach have shown an increased rate of relapse in patients who had not undergone RT. Our data were surprising that despite the lack of evidence for the omission of RT for Stage I and II HD, the present nationwide study revealed a >20% absolute decrease in the use of RT between 1988 and 2006. This large decrease might have resulted from patients not receiving centralized care in a multidisciplinary setting. It is paramount that all cases of early-stage Hodgkin’s lymphoma be considered for combined modality therapy. Treatment planning involving radiation oncologists and medical oncologists should be the standard of care.

Second malignancies are the most serious late complication after successful treatment of Hodgkin’s lymphoma. Conflicting reports have been published regarding whether the addition of RT to chemotherapy leads to a significant increase in the incidence of second malignancies compared with chemotherapy-alone strategies. Our study revealed that the actuarial risk of developing a second malignancy was equivalent between patients who had undergone RT vs. those who had not undergone RT. We had a greater number of secondary leukemia patients in the group who did not undergo RT, which might have been because this group had received more chemotherapy cycles. Also, older chemotherapy regimens such as mechlor-ethamine, vincristine, procarbazine, and prednisone, which used greater doses of alkylating agents could also explain the increased incidence of leukemia observed in this cohort. Several advancements have occurred in the treatment of HD during the past 20 years, including the omission of alkylating agents from multiagent chemotherapy and the use of involved-field RT. Furthermore, we demonstrated  that a significant decrease had occurred in the use of RT during the past 20 years. However, despite these changes, no significant difference was found in the actuarial rate of second malignancy development between those patients treated in an earlier treatment era between 1988 and 1997 and those treated between 1998 and 2007.

The present study was limited primarily by the information available from the SEER database. First, no information on RT technique (e.g., total dose, fraction size, beam energy) was available. Second, we could not determine which patients had unfavorable risk factors, such as bulky disease, three or more involved nodal regions, or an elevated erythrocyte sedimentation rate; however, we did adjust for all available patient and tumor characteristics. We also could not comment on what types of chemotherapy the patients had received. It is possible that patients treated in earlier years received less efficacious chemotherapy regimens. However, despite these limitations, we were able to show a benefit from RT after adjusting for the known patient and tumor characteristics. Furthermore, in an attempt to exclude patients whose performance status or other factors might have limited them from receiving a form of definitive therapy, we examined patients who had survived >1 year. In that cohort of 1-year survivors, we were also able to demonstrate a survival benefit for RT.

Conclusions 

The present study is one of the largest studies to examine the role of RT in the initial management of Stage I and II HD. Our analysis revealed a survival benefit with the addition of RT with no increase in the development of secondary malignancies. During the same period, the use of the RT for Stage I and II HD decreased by >20%.