Trend Analysis of Clinical Outcomes in Hodgkin Lymphoma over the Last Three Decades: Report from the Nebraska Lymphoma Study Group

BACKGROUND: With improvements in the treatment of Hodgkin lymphoma (HL) over time, we hypothesized a reduction in disease progression, and an increase in progression-free (PFS) and overall survival (OS) over the past 3 decades.

METHODS: This is a retrospective study that included previously untreated HL patients reported to the Nebraska Lymphoma Study Group between 1983 and 2012. To analyze trends in outcomes, we divided the cohort into 3 time periods of 10 years each (1983-1992, 1993-2002, 2003-2012) to represent early (E), middle (M) and late (L) time periods. Patient, disease, and treatment-related factors as well as clinical outcomes (cumulative incidence of progression, PFS and OS) were compared according to time period. Multivariate analyses (MVA) were performed using Cox regression to adjust for significant covariates.

RESULTS: In this retrospective review, 528 patients were identified during the study period: E (n=120), M (n=208), L (n=200). Patients in the more recent cohort (L) were more likely to be younger (p=0.008), have better performance score (p<0.001), have more sites of nodal involvement (p=0.05) and were more likely to receive ABVD (doxorubicin, bleomycin, vinblastine, dacarbazine) or Stanford V (mechlorethamine, doxorubicin, vinblastine, vincristine, bleomycin, etoposide, prednisone) as initial therapy (p<0.0001). In univariate analysis, there was a decrease in the 5-year cumulative incidence rate of disease progression over time (E=27% vs M=22% vs L=21%), and an increase in the 5-year probability of PFS (E=58% vs M=66% vs L=74%, p=0.004) and 5-year OS (E=67% vs M=80% vs L=85%, p=0.001). In MVA, improvement in the cumulative incidence of disease progression, PFS and OS over time was mainly driven by prognostic patient-related (age, sex) and treatment-related factors (type of chemotherapy regimen). Independent predictors of shorter PFS included age ≥45 y and male gender (Hazard Ratio [HR] 2.00 [95% CI 1.52-2.62, p<0.001] and 1.43 [1.09-1.87, p<0.01], respectively), while patients who received ABVD and Stanford V regimens as initial therapy experienced improved PFS (HR 0.39 [0.23-0.65, p<0.001] and 0.42 [0.28-0.64, p<0.001], respectively versus MOPP-like regimens). For OS, patients who were ≥45 y and males were more at risk of death (HR 4.34 [3.08-6.08, p<0.001] and 1.73 [1.26-2.37, p<0.001], respectively) while patients who received ABVD and Stanford V regimens as initial therapy were less likely to die (HR 0.36 [0.19-0.69, p <0.01] and 0.39 [0.23-0.65, p<0.001], respectively compared to MOPP-like regimens). Similarly, in MVA, a decrease in risk of progression was noted for patients who received ABVD and Stanford V regimens (HR 0.36 [0.19-0.69, p<0.01] and 0.39 [0.23-0.65, p<0.001], respectively) compared to MOPP-like regimens, while increased for patients who were ≥45 y and male (HR 4.33 [3.08-6.08, p<0.001] and 1.73 [1.26-3.37, p<0.001], respectively).

CONCLUSIONS: Changes in upfront treatment regimens for HL over the last 3 decades have resulted in significant improvement in PFS and OS, and modest improvement in the risk for progression.