Prognostic Significance of Bone Marrow Involvement in Hodgkin Lymphoma

Background:

The utility of bone marrow biopsy (BMB) in patients with Hodgkin lymphoma (HL) has been a controversial topic. Clinical stage 4 (CS4) has been shown to be an independent poor prognostic marker and is also included in the International Prognostic Scoring (IPS) index for predicting both progression-free survival (PFS) and overall survival (OS). The current guidelines still recommend performing a BMB in patients with stage IIB (with unfavorable risk factors), III and IV HL, mainly for staging purposes. Though bone marrow involvement (BMI) upstages the HL to stage IV, the prognostic significance of BMI remains unclear. The study was aimed to determine the prognostic significance of BMI in underserved patients with HL and to determine the prognostic importance of other blood parameters.

Methods:

The study was conducted at John H. Stroger Hospital of Cook County, an inner city tertiary care hospital providing care to the underserved population of Chicago. Charts of 241 patients diagnosed with HL were screened from tumor registry. Patients with incomplete charts were not included in the study. Socio-demographic, clinical and pathologic factors were recorded at the time of diagnosis. For comparative purpose, CS4 disease did not include patients with BMI. Kaplan-Meier and bivariate analyses were performed. Cox regression analyses were conducted to explore predictors of OS and PFS. Hazard ratios (HR) with 95% confidence intervals (CI) were obtained.

Results:

The study included 192 patients of which 41% were Afro-Americans, 34% were Hispanics and 21% were Caucasians. Median age was 34 years with 25.5% patients being older than 45 years and 68% patients being women. Seventeen percent patients were positive for HIV. Nodular sclerosis was the most common histologic subtype (55%), bulky disease was recorded in 19% patients and 61.5% patients had B-symptoms. CS4 disease was seen in 12% patients while 28% patients were stage III, 47% were stage II and 13% were stage I. Single-site BMB was done in 96% patients. BMB was positive for involvement with HL in 19% patients (n=37). Out of these 37 patients, 84% (n=31) had advanced stage (III & IV) HL. BMI was seen in 5% patients with early-stage HL (stages I-II) and 41% patients with advanced-stage HL. Median IPS score was 2 (range 0-6). Median values for clinical factors were: hemoglobin-11.8 g/dL, platelets 314.5 x103/uL, leukocytes 8.3 x103/uL, neutrophils 6x103/uL, lymphocytes 1.2 x103/uL and albumin 3.7 g/dL. Mean OS was 143 months (95% CI 126-160) with 5-year OS of 89%. Significant correlates of OS included: age 45 years or older (HR 2.83, 95%CI 1.25-6.43, P =0.013), HIV (HR 2.80, 95%CI 1.19-6.61, P =0.019), nodular sclerosistype (HR 0.29, 95%CI 0.12-0.71, P =0.006), CS4 disease (HR 3.05, 95%CI 1.20-7.77, P =0.019), BM positive for involvement with HL (HR 5.76, 95%CI 2.56-12.98, P <0.001), IPS score (HR 1.57, 95%CI 1.20-2.06, P =0.001), hemoglobin <10.5 g/dL (HR 3.57, 95%CI 1.56-8.19, P =0.003), platelets <150 x103/uL (HR 5.32, 95%CI 2.19-12.96, P <0.001) and lymphocytes <0.6 x103/uL (HR 2.97, 95%CI 1.21-7.28, P =0.017). Gender, albumin level, leukocytosis (>15 x103/uL) and relative lymphopenia (<8%) did not have a significant association with OS. When adjusted for CS4 disease, BMI remained an independent predictor of OS (HR 2.17, 95%CI 1.04-4.55, P =0.039). Mean PFS was 105 months (95% CI 97-117) with 5-year PFS of 78%. BMI (HR 2.38, 95%CI 1.21-4.67, P =0.012), albumin level (HR 0.71, 95%CI 0.53-0.94, P =0.017) and albumin-globulin ratio (HR 0.41, 95%CI 0.18-0.93, P =0.032) predicted PFS. None of the other factors had a significant association with PFS. BMI had inverse association with OS (mean OS 97.5 months vs. 164 months, P <0.001; 5-year OS 70% vs. 93.5%, P<0.001) and PFS (mean PFS 74 months vs. 112 months, P = 0.004; 5-year PFS 65% vs. 81%, P=0.030).

Conclusion:

BMI in HL indicates disseminated disease. Our study shows that BMI in patients with HL has a significant inverse relation with OS and PFS, independent of stage. Our data indicates that in patients with HL important prognostic information can be achieved by demonstrating BMI and it should be considered for inclusion in the IPS score. The rate of BMI is higher in our cohort, for both early and advanced stage HL, when compared with literature and maybe a limitation of our retrospective study. However our results still warrant that long term prospective studies should be designed to explore the prognostic significance of BMI.