Intracranial germ cell tumors: a single-institution experience

Intracranial germ cell tumors [GCTs] are not a common disease. We reviewed the experience of a single institution to determine the variables that affect treatment outcome. A retrospective review of patients with the diagnosis of intracranial germ cell tumors treated in a single institution [KFSHRC] during the period from March 1985 to December 2007. Fifty-seven patients with the diagnosis of intracranial GCT were recorded in the KFSHRC Tumor Registry during the period from 1985 to 2007. Seven patients with a pineal region tumor treated as germinomas in the earlier years without a tissue diagnosis were excluded. This retrospective study was restricted to the remaining 50 patients with a tissue or marker diagnosis: 31 germinomas and 19 non-germinomatous germ cell tumors [NGGCTs]. The 10-year overall survival [OS], event-free survival [EFS] and relapse-free survival [RFS] were 87%, 88% and 96% for patients with germinoma, with a median follow-up of 4.5 [range 2-17] years, compared with 26%, 29% and 46% for patients with NGGCT with a median follow-up of 3 [range 1.5-13] years. For NGGCT, variables favorably influencing OS were younger age [< 16 y vs >/= 16 y, P=.01], higher radiation dose [>50 Gy vs 1990 vs <1990 P=.002]. Tissue diagnosis of GCTs is mandatory prior to treatment except for patients with elevated markers. In germinoma, localized radiotherapy [RT] for M0 patients may be adequate. Long-term follow-up is needed to define the benefit of adding chemotherapy. For NGGCT, the use of combined modality treatment and RT dose >50 Gy are important factors that influence the outcome. Second-look surgery and resection of residual/refractory tumors is always recommended

Improved survival with combined chemo-radiotherapy in primary central nervous system lymphoma

Primary CNS lymphoma [PCNSL] is an aggressive primary brain tumor. Cranial irradiation alone rarely results in long term disease control or prolonged survival. We retrospectively analyzed data on the effect of adding high-dose methotrexate [HDMTX] prior to whole brain irradiation [WBI]. All patients with PCNSL diagnosed and managed during 1991-2004 were identified and demographic characteristics, prognostic factors, treatment and outcome were reviewed. Of 62 patients, 10 were excluded [4 had WBI < 40 Gy and 6 had no treatment]. Radiation alone was considered curative with a dose > 40 Gy. Combined modality therapy included 3-4 cycles of HDMTX [3g/m2] followed by WBI. Of 52 patients analyzed for outcome, 36 had WBI [dose > 40 Gy], 16 received 3-4 cycles of HDMTX followed by WBI [combined modality therapy [CMT]]. Median age was 48.2 years; 42 years in the CMT group, 51 years in WBI. Patient characteristics were comparable between two groups except for higher multifocal tumor in the CMT group [92% vs. x22%, P=.029]. Median follow up was 12.83 +/- 6.4 months. The hazard ration for an event was 0.64 [95% CI, 0.52-0.98] and for death 0.58 [95% CI, 0.48-0.92], both in favor of CMT. Univariate regression analysis using one-way analyses of variance [ANOVA] and multivariate Cox regression analysis for prognostic factors including age [< 60 vs. >60], ECOG PS [0-2 vs. 3-4], extent of surgery [biopsy vs. debulking], solitary vs multifocal tumor and dose of radiation therapy [>50Gy vs. >50 Gy] failed to identify any prognostic factor. This retrospecitive comparison supports phase II trial results that indicate that high-dose methotrexate followed by WBI in PCNSL improves outcome

Morphologic, immunphenotypic and clinical discriminators between T-cell / histiocyte-rich large B-cell lymphoma and lymphocyte-predominant Hodgkin lymphoma

Features of T-cell/histiocyte rich large B-cell lymphoma [THRLBCL] overlap with those of lymphocyte predominant Hodgkin lymphoma [LPHL]. The two lymphomas may represent a spectrum of the same disease, and differentiation between the two can sometimes be difficult. We looked at histomorphologic, immunophenotypic and clinical information that may help differentiate the two entities. Cases of THRLBCL and LPHL were blindly reviewed and studied for histological pattern [nodular vs. diffuse], nuclear features and pattern of expression of CD20, CD30, CD57, epithelial membrane antigen [EMA] and Epstein-Barr virus [EBV]. A score encompassing diffuse histology, high nuclear grade, CD20 single-cell pattern, CD30+, CD57-, EMA-, and EBV+ was estimated for the diagnosis of TCHRLBCL. There were 58 cases, including 30 cases of TCHRLBL and 28 cases of LPHL. The median age was 36 years for TCHRLBCL and 21 years for LPHL [P=0.0001]. Three types of nuclei were identified [lymphocytic/histocytic, Reed-Sternberg and centroblast-like]. The latter two high-grade nuclei were suggestive of TCHRLBCL. TCHRLBCL and LPHL, respectively, showed diffuse histology, 90% vs. 4% [P=0.001], single CD20+ cells, 93% vs. 3.5% [P=0.00004], CD30+ cells, 30% vs. 0% [P=0.01], CD57+ cells, 41% vs. 93% [P=0.008], EMA+ cells, 27% vs. 60% [P=0.113], EBV+ cells, 24% vs. 0% [P=0.117], high nuclear grade, 70% vs. 0% [P=0.001], total score 2-7 [mean 4.68] vs. 0-2 [mean 0.72] [P=0.001], high stage, 86% vs. 7% [P=0.0001]. Our findings indicate that a combination of multiple parameters can help differentiate between the two diseases. Two cases originally diagnosed as LPHL were re-assigned the diagnosis of THRLBCL