Supradiaphragmatic Consolidative Radiation Therapy in the Management of Stage III Hodgkin's Disease Treated with ABVD-Based Chemotherapy.

ABVD-based chemotherapy alone is a standard treatment for stage III Hodgkin's disease. The role for radiotherapy in this patient population remains controversial. In this study, we evaluated factors such as initially involved site or size of disease that could predict for local recurrence and thus may guide consolidative radiotherapy use and design of future prospective studies.

We retrospectively reviewed the medical records of 118 Stage III Hodgkin's Disease patients diagnosed and treated at the University of Texas M.D. Anderson Cancer Center from 1993-2006. We evaluated patterns of failure, site and size of initial involvement, image-verified bulky disease, and site-specific consolidative RT on rate and site of recurrence and survival. We defined local failure (LF) as failure at initial site of involvement, loco-regional failure (LRF) as any failure above the diaphragm and freedom from failure (FFF) as any failure event, including below diaphragm and extralymphatic spread. We used descriptive statistics to summarize the demographic and clinical characteristics of the patients. We used the Kaplan-Meier method to estimate LRF, FFF, disease free survival (DFS) and overall survival (OS) and comparisons of potential prognostic factors by log-rank tests. Chi-square analysis was used to compare differences in proportions for LF. Multivariate analysis was performed using Cox proportional hazards regression model to determine prognostic factors that contributed to prognosis.

Patient demographics were: median age 36 years (range 17-78 years), female 42.9%, B symptoms 44.4%, stage “3S” 19.4%, nodular sclerosis subtype 66.4%, and mixed cellularity subtype 18.5%. Initial sites of involvement were: axilla 44.5%, head and neck (H&N) 88.2%, mediastinum 81.5%, abdomen 89.9%, pelvis 42.9%, groin 20.2%. The median follow-up for all patients was 67 months.

Of the 118 patients treated at M.D. Anderson, 88% achieved a complete response (CR). Of those achieving CR, 71% received ABVD 6 cycles or more, and 37.5% received consolidative RT. After achieving CR, 14/88 (13.5%) had failures above the diaphragm, 1/88 (1.1%) failed below the diaphragm, and 1/18 (1.1%) failed on both sides of diaphragm. Further evaluation of those that failed above the diaphragm demonstrated that the most common sites of failure were H&N 9/18 (50%), mediastinum 5/18 (27.8%), and axilla 3/18 (16.7%). Patients with initial axilla involvement were more likely to exhibit failure above the diaphragm after CR when compared to those without initial axilla involvement (5-Yr LRF: 23.6% involved vs. 3.6% uninvolved, p < 0.02). The same was also true when comparing those with and without initial mediastinal involvement (5-Yr LRF: 9.5% involved vs. 0% uninvolved, p = 0.05). For those with initial H&N involvement, no significant differences in LRF were observed. We did observe higher failure rates for those with bulky H&N disease when compared to non-bulky H&N (5-Yr LRF: 35.7% bulky vs. 5.7% non-bulky, p < 0.05).

The addition of mediastinal RT improved the 5-years LRF from 18% to 3.8% (p < 0.05), the 5-years FFF from 72.4% to 88.9% (p < 0.05), and 5-years DFS from 82.4% to 93% (p = 0.002). The addition of H&N RT showed a trend toward both improved LRF (5-Yr: Yes RT 0% vs. No RT 16.3%, p 0.091) and DFS (5-Yr: Yes RT 94.4% vs. No RT 86%, p = 0.068). In multivariate analysis, prognostic factors significant for increased local failure events were initial axilla and/or mediastinal involvement and bulky H&N disease. The lack of B symptoms and more than 6 cycles of ABVD were associated with improved DFS (p <0.01). Lack of mediastinal RT had a worse DFS (p=0.02). For OS, younger age and more than 6 cycles of ABVD were associated with improved survival with (p <0.05, and <0.001, respectively), whereas initial mediastinal involvement or mediastinal failure correlated with a worse outcome (p < 0.01). Supradiaphragmatic RT was not associated with an increased risk of second malignancies or heart disease.

For patients with stage III Hodgkin's, disease below the diaphragm appears to be well managed by chemotherapy alone (at least when 6 cycles of ABVD is given). For disease above the diaphragm, our limited study suggests that consolidative RT after CR appears to confer a benefit to those with initial axilla or mediastinal involvement, or bulky H&N disease. A randomized trial further exploring these findings would be beneficial.

No relevant conflicts of interest to declare.