Habermann, Thomas M.

Abstract Background: B-cell non-Hodgkin lymphomas (NHL) are common lymphoid malignancies in which malignant B-cells arrested at various stages of differentiation proliferate within lymph nodes and occasionally other tissues. The microenvironment that supports the growth and survival of NHL B cells consists of a complex network of immune cells and cytokines and its specific composition has been shown to have significant clinical implications. The intratumoral immune cells include effector and regulatory T (Treg) lymphocytes that are more than simple residual elements from the normal lymph node structure. We previously explored whether the extent of T cell infiltration played a role in the clinical outcome of patients with B-cell NHL and found that an increased percentage of CD4+ T cells in the diagnostic biopsy of lymphoma patients significantly correlated with an improved 5-year overall survival. We have also shown that intratumoral Treg cells significantly suppress the anti-tumor response. Based on the fact that intratumoral T-cells are important in B-cell NHL, we evaluated whether genetic variation in genes that regulate the T-cell response may be associated with lymphoma risk. Methods: We genotyped 257 single nucleotide polymorphisms (SNPs) from 50 candidate genes related to T-cell differentiation and function in a clinic-based study of 441 Caucasian NHL cases and 475 frequency matched Caucasian controls seen at the Mayo Clinic from 2002–2005. Tagging and nsSNPs were selected from HapMap. The most prevalent homozygous genotype was used as the reference group and each SNP was modeled individually as having a log-additive effect, expressed as ordinal odds ratios (OR) per variant allele and 95% confidence intervals, in an age- and sex-adjusted logistic regression analysis. We also evaluated the consistency of the findings for the most common types of B-cell NHL - diffuse large B-cell lymphoma (DLBCL), follicular lymphoma (FL), and CLL/SLL. For gene-level analyses, we used principal components (PC) analysis using all SNPs from a gene; those PCs that explained > 90% of the variability were then modeled in a logistic regression analysis and significance was determined using a likelihood ratio test. We also evaluated the association of haplotypes from each gene with risk of NHL using the score test implemented from HAPLO.SCORE. Results. The mean age at diagnosis was 60.1 years for cases and 58% were male; among controls, the mean age at enrollment was 61.7 years and 55% were men. In the PC gene analyses, PRF1 (perforin gene - involved in cytotoxicity; p=0.004), CD276 (B7-H3 gene - involved in co-stimulation; p=0.01), TBX21 (T-bet gene – regulates Th1 cell development; p=0.02), and IL6 (p=0.02) were significant at p<0.05; haplotype analysis showed similar results, with the addition of CARD15 (NOD2 gene – regulates Th1 responses; p=0.02). There were 2 SNPs in PRF1, and both the intronic SNP rs3758562 (OR=1.38; 1.13–1.69) and the synonymous coding SNP rs885821 (OR=0.80; 0.63–1.02) were associated with NHL risk overall and for each subtype. There were 2 SNPs in CD276, and both the intronic SNP rs7176654 (OR=1.29; 1.07–1.56) and the mrna-utr SNP rs3816661 (OR=0.82; 0.68–0.99) were associated with NHL risk overall and with the CLL/SLL and follicular NHL subtypes specifically. There were 2 SNPs in TBX21, and only the mrna-utr SNP rs7502875 (OR=0.73; 0.59–0.92) was associated with NHL risk overall and for each subtype. There were 11 SNPs in IL6, and the intronic SNP rs2069835 (OR=1.88; 1.24–2.84) and the locus-region SNP rs2069824 (OR=1.73; 1.20–2.50) were each associated with NHL risk overall and each subtype. Finally, there were 9 SNPs in CARD15, with only the nonsynonymous coding SNP rs5743291 (OR=1.43; 1.02–1.98) being associated with overall NHL risk, and risk was specific to FL. Conclusions. Genetic variation in several genes that play critical roles in T-cell maturation and function was associated with lymphoma risk. These genes appeared to be associated with the differentiation and function of effector T-cells particularly Th1 and Th17 cells. These results provide potentially important insights into the role of T-cells in lymphomagenesis.

Abstract Background: Reactive oxygen species cause DNA damage and altered immunologic responses, and have been linked to the development of NHL. Diets rich in fruits and vegetables are excellent sources of antioxidants, and vitamins C, E, polyphenols, and carotenoids, along with selected micronutrients such as zinc are thought to be responsible for most of the antioxidant activity in foods. We evaluated the hypothesis that vegetables, fruits and nutrients involved in antioxidant activity would protect against developing NHL. In a secondary analysis, we also assessed heterogeneity of this hypothesis for the most common subtypes of NHL: chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL), follicular lymphoma, and diffuse large B cell lymphoma (DLBCL). Methods: We evaluated dietary antioxidant intake and NHL risk in a clinic-based study of 416 newly diagnosed NHL cases and 926 frequency-matched controls enrolled at the Mayo Clinic from 2002–2007. Usual diet two years before diagnosis/enrollment was assessed using a self-administered, 128-item food frequency questionnaire. Dietary intake of antioxidants was estimated using the Food Processor SQL system, further supplemented with data from US Department of Agriculture nutrient databases. Unconditional logistic regression was used to estimate odds ratios (ORs) and 95% confidence intervals (CI), adjusted for age, sex, residence, and total energy. NHL subtypes were centrally reviewed, and subtype-specific risks were estimated using polychotomous logistic regression. Results: The mean age at diagnosis was 60.8 years for cases and 57% were male; for controls, the mean age at enrollment was 60.8 years and 54% were men. NHL risk was inversely associated with intake of total vegetables (OR for highest compared to lowest quartile, 0.46; 95% CI 0.32–0.68; p-trend<0.001), and specifically green leafy (OR=0.54; 95% CI 0.39–0.76; p-trend<0.001) and cruciferous (OR=0.62; 95% CI 0.43–0.88; p-trend=0.02) vegetables, but not legumes or red/yellow/orange vegetables. There was no association with total fruit intake or intake of citrus fruits in particular. When modeled together, the association for green leafy vegetables was unchanged while the association for cruciferous vegetables attenuated (OR=0.75; 95% CI 0.50–1.13; p-trend=0.3). Higher intake of total vitamin A (OR=0.60; 95% CI 0.41–0.88; p-trend<0.001) and vitamin E (OR=0.53; 95% CI 0.34–0.82; p-trend<0.001), but not vitamin C, were associated with lower NHL risk. For vitamin A, there was no association with preformed Vitamin A, but there was an inverse association with beta-carotene (OR=0.56; 95% CI 0.39–0.81; p-trend<0.001); only weak inverse associations were observed for other carotenoids (lutein/zeaxanthin, lycopene, and cryptoxanthin). When modeled together, the inverse association for vitamin E was unchanged while the association for beta-carotene attenuated. For micronutrients, there were inverse associations for manganese (OR=0.60; 95% CI 0.40–0.90; p-trend=0.02), selenium (OR=0.47; 95% CI 0.27–0.80; p-trend=0.01), and zinc (OR=0.46; 95% CI 0.27–0.80; p-trend<0.001) but not copper; when modeled simultaneously, both manganese (p-trend=0.03) and zinc (p-trend=0.02) remained significantly and inversely associated with NHL risk. Further adjustment for education, family history of NHL, smoking, alcohol use, and body mass index did not alter these results. There was no NHL subtype heterogeneity for the associations with total vegetables, green leafy vegetables or vitamin E, while inverse associations were not observed for manganese and follicular lymphoma or zinc and CLL/SLL. Conclusion: Higher intakes of vegetables, particularly green leafy vegetables, vitamin E, manganese, and zinc were associated with lower risk of NHL. These finding extend a growing literature from case-control and cohort studies that support a protective role of foods and nutrients that function in antioxidant related pathways against development of NHL, and therefore may represent one of the few known modifiable risk factors for this cancer.

Abstract Introduction The role of body mass index (BMI) impacting clinical outcome among lymphoma patients is controversial. Two recent studies suggest that increased BMI is associated with significantly improved survival. In this study the association between BMI at study entry and failure-free survival (FFS) and overall survival (OS) was evaluated in three phase III Eastern Cooperative Oncology Group-led trials, among patients with DLBCL (E4494), follicular lymphoma (FL) (E1496) and Hodgkin's lymphoma (HL) (E2496). Patients and Methods 537 patients with DLBCL, 730 patients with HD and 282 patients with FL were included in the analysis. BMI was calculated as weight (kilograms) divided by the square of height (meters), using data at study entry. BMI was analyzed both as continuous and categorical variables (underweight <18.5 kg/m2, normal weight: 18.5 to < 25 kg/ m2, overweight: 25 to < 30 kg/ m2, and obese :≥ 30 kg/ m2).The underweight group was excluded due to low (< 2%) prevalence. Baseline patient and clinical characteristics, treatment received and clinical outcomes were compared across BMI categories. PFS was defined as the time from study entry to relapse, progression, or death. OS was measured from study entry to death of any cause. The log-rank test and Cox regression models was used to check the association. The association between BMI and FFS/OS was also independently assessed among patients treated with rituximab. A sensitivity analysis was performed excluding patients with significant weight loss at baseline. Results Among patients with DLBCL, HL and FL, the median age was 70, 33 and 56; 29%, 29% and 37% were obese and 38%, 27% and 37% were overweight, respectively. Age was significantly different among BMI groups in all three studies. Higher BMI groups tended to have better prognosis at study entry among DLBCL and HL patients. BMI was not associated with clinical outcome, with p-values of 0.89, 0.30 and 0.40 for FFS, and p-values of 0.64, 0.67 and 0.09 for OS, for patients with DLBCL, HL and FL, respectively (Figure 1). In multivariate analysis adjusting for other clinical factors, BMI remains an insignificant predictor for all three histologies (Table 1). When limiting to patients treated with rituximab, the association remains non-significant for both FL patients (p=0.92 for PFS, p=0.36 for OS) and DLBCL patients (Figure 2). A subset analysis of males with DLBCL treated on R-CHOP, which matched the study cohort used in a recent report (Carson et al, 2012), revealed no differences in FFS (p=0.48) or OS (p=0.58) (Figure 2). Sensitivity analysis excluding patients with weight loss at study entry resulted in similar findings. Conclusion BMI was not significantly associated with clinical outcomes among patients with DLBCL, HL or FL, in three prospective phase III clinical trials. The findings contradict some previous reports of similar investigations. Further work is required to understand the observed discrepancies. Disclosures: Horning: Genentech: Employment.

Abstract This study evaluated the contributing roles of flow cytometric immunophenotyping of blood and bone marrow and immunohistochemical paraffin section staining of bone marrow biopsies in the staging of B-cell malignant lymphoma. Flow immunophenotyping was performed on a marrow specimen in 175 cases; a corresponding blood specimen was also immunophenotyped in 135 of these cases. Morphologic marrow involvement by lymphoma was found in 59 cases; flow immunophenotyping identified 54 cases with a monoclonal B-cell process: morphology-positive/flow-positive (n = 49), morphology-positive/flow-negative (n = 10), morphology-negative/flow-positive (n = 5), and morphology-negative/flow-negative (n = 111). The 10 morphology-positive/flow-negative cases included 5 follicular and 5 large-cell lymphomas with minimal marrow involvement. All 5 morphology-negative/flow-positive cases were from patients with large-cell lymphomas and bulky clinical disease. Because the blood contained the same B-cell clone in 2 of 2 morphology-negative/flow-positive cases studied, blood contamination of marrow may account for these findings. Blood flow cytometric immunophenotyping studies were positive in 32 cases; 30 had marrow involvement by morphology and were from patients with follicular, mantle cell, lymphoplasmacytic, small lymphocytic, or marginal zone lymphomas. From our results, we conclude that (1) bone marrow flow cytometric immunophenotyping is not a cost-effective replacement for good morphologic evaluation in lymphoma staging and that (2) a positive peripheral blood flow cytometric immunophenotyping study when performed in low-grade lymphomas correlates with marrow involvement.

Abstract Abstract 94 Background: Lymphoma diagnosed during pregnancy is a rare occurrence. Further, there is heterogeneity in clinical presentation and a range of lymphoma histologies resulting in a continued deficiency of lymphoma-specific data in the literature regarding prognosis, optimal therapy, maternal complications, and fetal outcome. Methods: A comprehensive retrospective analysis was completed for patients (pts) diagnosed (dx) with Hodgkin lymphoma (HL) or non-Hodgkin lymphoma (NHL) during pregnancy at 9 large U.S. academic centers over a 13-year period (1998–2011). The vast majority of cases were co-managed with high-risk maternal fetal medicine. Data on maternal disease characteristics, treatment details, obstetric complications, and fetal outcomes were analyzed. Results: 88 cases were identified with complete data available in 82. The median age of pts was 31 yrs (18–40). Lymphoma dx occurred at a median of 24 weeks (wks) gestation (5–40). 15% of pts were dx during the 1st trimester, 46% during the 2nd trimester, 35% during the 3rd trimester, while 4% (n=3) of pts had a pre-existing NHL dx (follicular lymphoma (FL)). 52% (n=43) of pts had NHL and 48% (n=39) HL. Of NHL cases, 83% (n=33) were B-cell and 17% (n=10) T-cell. B-cell histologies were DLBCL n=23 (including 2 primary CNS and 1 'double hit' NHL), FL n=5 (including 1 leukemic phase FL), Burkitt's lymphoma n=3, and mixed histology n=1; T-cell histologies were anaplastic large cell n=6, PTCL not otherwise specified n=2, and T-cell/NK NHL n=2. Pts with B-cell NHL and HL were dx at a median age of 29 yrs (range, 18–40), while T-cell NHL pts were dx at a median of 34 years (19–37). MRI without gadolinium was the most common imaging modality utilized for staging. 63% of NHL and 19% of HL pts had stage III/IV disease (25% of HL with stage IIB). Pregnancy was terminated in 6 pts (n=4 aggressive NHL and n=2 HD) to enable immediate chemotherapy (five in the 1st trimester [at 5, 8, 10, 10, and 12 wks] and 1 pt in the early 2nd trimester who required high-dose methotrexate). 48 pts (63%) initiated anti-lymphoma therapy at a median of 25 wks gestation (range, 13–37) with 79% of pts initiating therapy in the 2nd trimester (see Table 1). The median gestation at time of lymphoma dx for pts who received intra-partum treatment was 20 wks vs 34 wks for pts who had therapy deferred to post-delivery (p<0.0001). The overall response rate among pts who received therapy was 87% (74% complete remission). Notably, gestation went to full-term in 54% with delivery occurring at a median of 37 wks (31–40). Therapy was deferred until post-delivery in 28 pts (37%) and the majority of these pregnancies were also carried to term (median delivery at 38 wks). At a median follow-up of 41 months (6–157), the 3-year PFS and OS for all pts were 79% and 89% respectively. Histology-specific 3-year PFS and OS were: B-cell NHL: 73% and 82%, respectively; T-cell NHL: 50% and 90%, respectively; and HL: 90% and 95%, respectively (Figure 1). Detailed obstetrical information was available in 59 pts. There were minimal pre-term complications, the most common being induction of labor in 45%. Perinatal events included spontaneous rupture of membranes in 5% and pre-eclampsia in 8%. Further, there were no differences in these events detected among pts who received intra-partum treatment vs deferred therapy. No episodes of chorioamnionitis or endometritis were noted. There was 1 stillbirth that occurred at 19 wks gestation in a 34-year-old pt with double-hit NHL following 1 cycle of R-CHOP. The median birth weight of infants was 2427 grams (1005–5262) with no difference among pts who received intra-partum chemotherapy vs not (2637 grams vs 2212 grams, respectively; p=NS). Microcephaly was reported in 1 case following 4 intra-partum cycles of CHOP for DLBCL; there were otherwise no malformations detected. Conclusions: To our knowledge, this represents one of the largest experiences reported of lymphoma in pregnancy. Our data show that standard NHL and HL chemotherapeutic regimens (without anti-metabolites) administered during the 2nd and 3rd trimester, including as early as 13 wks gestation in select cases, is associated with minimal maternal complications or fetal detriment. In addition, pts with low risk clinical scenarios (e.g., indolent NHL and/or late gestational dx) had therapy safely deferred to post-partum. In our experience, this approach was associated with overall expected lymphoma-related survival. Disclosures: No relevant conflicts of interest to declare.

Abstract Abstract 85 Background: Follicular lymphoma (FL) is an indolent non-Hodgkin lymphoma (NHL) with a highly variable natural history. Clinical variables describing either patient- or tumor-specific characteristics, many of which are included in the Follicular Lymphoma International Prognostic Index (FLIPI), are used to risk-stratify patients at the time of diagnosis. However, characteristics of non-malignant cells present within the tumor microenvironment provide important prognostic information. For example, both immunohistochemistry and gene expression profiling studies demonstrate that lymphoma-associated macrophages (LAMs) are prognostically important in FL. As LAMs are monocyte-derived, we sought to determine whether the absolute monocyte count (AMC) obtained from a CBC at diagnosis is an adverse prognostic factor in FL. Methods: Consecutive patients with FL who were evaluated and treated at the Mayo Clinic, Rochester between 1998 and 2007 were considered for study participation. Study approval was granted by the Mayo Clinic Institutional Review Board. The primary objective of the study was to determine the significance of the AMC, obtained at the time of diagnosis from a standard complete blood count with differential, in predicting overall survival. The secondary study objective was to determine whether the AMC is an independently significant prognostic factor when compared with the FLIPI. Results: The median follow-up was 4.2 years (range: 2 months to 11 years) for the 355 patients included in this study. At diagnosis, 43%, 34% and approximately 23% of patients were classified as low-, intermediate- or high-risk by the FLIPI. The median AMC at diagnosis was 510/μL, with 8% of patients presenting with an absolute monocytosis. At diagnosis, 45% of patients were initially observed, 42% received combination chemotherapy or immunochemotherapy alone, while the remaining 13% underwent surgery or radiation. The AMC was evaluated as both a continuous and dichotomized variable on univariate analysis. The optimal cut-point for the AMC was determined to be 570/μL by ROC analysis and was used when analyzing the AMC as a dichotomized variable. A relative elevation in the AMC was associated with inferior overall survival, whether analyzed as a continuous or dichotomized variable, with hazard ratios (HR) of 1.002 (HR per unit change in AMC; 95% confidence interval 1.001–1.002; p=0.0002) and 3.4 (95% confidence interval 2.0–6.0; p<0.0001), respectively. The median OS for patients with an AMC <570/μL at diagnosis has not been reached, whereas a median OS of 10.2 years (95% confidence interval 7.8–11.2 years; p<0.0001) was observed for those patients with an AMC ≥570/μL. For comparison, median OS has not been reached for high-risk patients identified by the FLIPI. Among those patients who were initially observed, the median time to progression (TTP) was 31 months (95% confidence interval 25–37 months) for those with an AMC <570/μL at diagnosis. In contrast, patients with an AMC ≥570/μL experienced a significantly shorter TTP following initial observation, with a median TTP of 18 months (95% confidence interval 8–28 months; p=0.03). When adjusting for prognostic factors included in the FLIPI on multivariate analysis, the association between the AMC as a dichotomized variable and survival remained significant with a HR of 2.6 (95% confidence interval 1.5–4.6; p=0.001). When the FLIPI alone (high vs. low/intermediate) was analyzed on multivariate analysis, the AMC remained an independent predictor of survival (p<0.0001). Furthermore, the AMC provided additional prognostic information when superimposed upon the FLIPI. For example, a subset of patients classified as high-risk by the FLIPI were observed to have an OS comparable to low/intermediate risk patients identified by the FLIPI upon further risk-stratification by the AMC, with a 6-year OS of 80%. In contrast, high-risk FLIPI patients with an AMC ≥570/uL were observed to have a median OS of 4.2 years (95% confidence interval 2.7–7.2) and 6-year OS of 47%. Conclusions: The AMC is associated with overall survival in FL and is independent from the FLIPI on multivariate analysis. While the AMC did not appear inferior to the FLIPI in identifying high-risk patients, it may be most helpful when used in conjunction with the FLIPI. These results further support the central role of nonneoplastic myeloid-lineage cells in FL biology. Disclosures: No relevant conflicts of interest to declare.

Abstract Abstract 879 Background: Several solid organ transplant (SOT) PTLD series have reported that PCNS disease is a poor prognostic factor (e.g., Leblond et al, JCO 2001; Maeker et al, JCO 2007; Knight et al JCO 2009; Evens et al, JCO 2010). However, these reports contained small PCNS PTLD numbers and most series studied patients (pts) in the pre-rituximab era. Methods: We performed an International retrospective analysis of PCNS SOT-related PTLD diagnosed from 1998 through 2010. Cases with secondary CNS localization were excluded. Detailed pt and disease characteristics, including frontline therapy, were evaluated. Fisher's exact test was used to compare PCNS PTLD vs non-PCNS PTLD cases. Further, univariate associations with survival were determined, while a multivariate Cox regression proportional hazards model was performed. Results: 71 PCNS PTLD cases were identified, of which 65 had complete data. There were 35 men and 30 women; the median age was 53 years (range, 20–77 years). The most common type of SOT was kidney in 72%, followed by lung 18% and 5% each pancreas and heart. The median time from SOT to PCNS PTLD was 46 months (range, 2 to 504 months). Further, only 18% occurred within 12 months of SOT, and 32% of PTLD diagnoses occurred > 10 years after SOT. 85% of pts had received prior mycophenolate, while 30% received tacrolimus prior to PCNS PTLD. According to WHO histology, 83% of PCNS PTLD cases were monomorphic subtype (all B-cell), while tumor was EBV+ in 93% despite EBV being detectable in the peripheral blood (by PCR) in only 38% of these cases. In addition, 49% of pts had poor performance status (i.e., 2–4), 48% had elevated LDH, and 48% were hypoalbuminemic at diagnosis. Interestingly, several of these characteristics were notably different when compared with a non-PCNS PTLD cohort (Table 1A). All pts here presented with CNS-related symptoms and 98% had abnormal findings on MRI showing a parenchymal brain lesion(s); 50% of pts had >1 CNS lesion and 40% had involvement of deep regions of the brain. Further, 12% had concomitant CSF involvement, while no pts had ocular involvement. In terms of therapy, 89% of pts had reduction in immunosuppression (RI). Additional frontline therapy consisted of high dose methotrexate (HD-MTX)-based therapy in 43% (46% of whom also received concurrent rituximab), whole-brain radiation in 20% (61% with rituxmab), high-dose cytarabine (HDAC)-based in 15% (30% with rituximab), while 11% received rituximab alone and 11% had no other therapy (i.e., RI alone). Altogether, 48% of pts received rituximab as part of initial therapy. The overall response rate (ORR) was 55% with 30% complete remission (CR) rate. With a median follow-up of 39 months (1–135), the 3-year progression-free survival (PFS) and overall survival (OS) for all pts were 31% and 32%, respectively (Figure 1). Pts who received rituximab vs not as a component of first line therapy had superior PFS (40% vs 22%, respectively, p=0.02) and borderline improved OS (43% vs 22%, respectively, p=0.06), while pts who did not receive rituximab nor HD-MTX-based therapy as frontline therapy did particularly poor (PFS and OS both 10%). Several pre-treatment and treatment-related factors predicted survival on univariate analysis (Table 1B). When analyzing only pre-treatment related factors in the Cox multivariate regression model, only increasing age remained significant (PFS: HR 1.04 [95%CI 1.00–1.07], p=0.031; and OS: HR 1.04 [95%CI 1.01–1.08], p=0.020). With treatment-related factors included in the model, response to frontline therapy (CR or partial remission) was the dominant predictive factor of survival on multivariate analysis (PFS: HR 0.136 [95%CI 0.047–0.390], p=0.0002; and OS: HR 0.130 [95%CI 0.045–0.381], p=0.0002). Conclusion: These data represent the largest report of PCNS PTLD to date. PCNS PTLD appears to represent a distinct biologic and clinical entity within the PTLD spectrum that retains EBV positivity, occurs late, and presents with additional high-risk features. Furthermore, outcome of PCNS PTLD remains overall modest. However, these data suggest that rituximab and HD-MTX appear to favorably impact survival. Continued study of this subset of pts is critically needed. Disclosures: No relevant conflicts of interest to declare.

Abstract Abstract 880 Pulmonary marginal zone lymphoma is a distinct subgroup of B-cell non-Hodgkin's lymphomas originating from the bronchial-associated lymphoid tissue (BALT) and characterized by an indolent clinical course. Here we present the data of the IELSG-28 study of the International Extranodal Lymphoma Study Group (IELSG) on primary pulmonary lymphoma patients. Clinical records and histological material of 186 patients diagnosed with pulmonary marginal zone lymphoma from December 1986 to June 2009 were collected. The pathology was reviewed centrally and the diagnosis was confirmed in 180 (96%) patients. The median age at diagnosis was 62 (30 – 88). The majority of the patients (149/180) had localized disease at diagnosis with a good performance status (0–1) and low prognostic index (IPI 0–2). Eighty-one patients (45%) received local treatment, including surgery (74%) or radiotherapy (4%). Ninety-nine patients with advanced stage disease or incomplete surgical resection received an additional systemic treatment. Most of these patients (66; 66%) received an alkylating containing regimen, 22 (22%) patients were treated with an anthracycline containing regimen and 29 (29%) received the anti-CD20 monoclonal antibody in combination with chemotherapy. With a median follow up of 67 (range 1–256) months, 47/180 patients (26%) experienced disease progression. (median time to progression 30 months, range 0–109). Progression free survival analysis showed a trend in favour of patients treated with immuno-chemotherapy versus those receiving chemotherapy alone (HR 0.53–95%CI: 0.21, 1.33), while no difference in terms of OS and PFS was observed between patients receiving systemic anthracycline or alkylating containing regimens. In twenty five cases a specific PCR-assay was carried out for detecting the presence of Achromobacter (Alcaligenes) xylosoxidans. The presence of Achromobacter was detected in 50% of patients. Our results confirm that local therapy (surgery or radiotherapy) is the treatment of choice for patients with localized disease. Chemotherapy has to be considered in case of relapse, incomplete surgical excision or advanced disease. In this setting, alkylating containing regimens plus rituximab is the most effective option. Further studies are ongoing to confirm the relevance of infection/colonization with Achromobacter (Alcaligenes) xylosoxidans for pathogenesis of BALT lymphoma. Disclosures: No relevant conflicts of interest to declare.

Abstract Abstract 87 Background: The International Prognostic Index (IPI) remains the most powerful predictor of clinical outcome in DLBCL in the R-CHOP era, serving as a useful surrogate for the biology we are just beginning to understand. With a follow-up of more than 9 years, its uniformly staged and treated patients and its prospectively collected unstained slides for correlative studies, the clinical data set from E4494, the US Intergroup trial comparing CHOP and RCHOP in patients >60 years with DLBCL, stands as a valuable resource for investigating prognosis in DLBCL. Based on immunohistochemistry, we previously showed that rituximab modulates the prognostic significance of some biomarkers in DLBCL. With these same specimens, the quantitative nuclease protection assay (qNPA), a methodology for measuring mRNA levels in FFPET, was used to develop separate prognostic signatures for the CHOP and RCHOP arms that can be simply applied using stored unstained slides. Methods: Five micron unstained FFPET sections from 183 eligible and evaluable cases enrolled on E4494 and submitted for prospective immunohistochemical correlative studies more than ten years ago were used for this analysis. Tissue was scraped from slides and a multiplexed qNPA was performed in triplicate using a customized Array Plate assay (HTG, Inc) for 43 genes of interest. TBP served as a housekeeping gene. Association between standardized log gene expression and patient failure-free survival (FFS) and overall survival (OS) was obtained using the Cox proportional hazards model. A weighted analysis was used to eliminate the confounding effect of maintenance rituximab. Genes that showed at least marginal significance in the univariate analysis were used to perform LASSO (penalized method to select best subset) to select a final list of genes in the multivariate analysis. Using the predictive model for either CHOP or R-CHOP induction, individual risk scores were calculated based on the multivariable model, and cases were dichotomized into low and high risk groups based on the median risk score. The model was then validated using the Lenz dataset (NEJM, 2008). Results: In six cases, tissue from slides prepared >10 years ago was compared to freshly cut sections from corresponding blocks, and showed excellent concordance (corr=0.86). On-study characteristics for the 176 cases with analyzable data were representative of the greater E4494 patient population. Six gene predictors were developed for each arm of the trial: RCHOP: FN1, LMO2, AKT1, HIF1a, AKT3, BCL2; and CHOP: PDCD4, HLADRB1, COL3A1, LMO2, ROBO4, TP53. Both signatures proved powerful predictors of FFS and OS among CHOP (FFS: p=0.0031; OS: p=0.0013) and RCHOP (FFS: p=0.001; OS=p=0.0015) treated patients. When adjusted for the clinically-based IPI, the gene-risk score retained its significance (CHOP: FFS p=0.0007; OS p=0.0011; RCHOP: FFS p=0.003; OS, p=0.001) while the IPI became only marginally significant (CHOP: FFS: p=0.06, OS p=0.06; R-CHOP: FFS p=0.09, OS p=0.02), suggesting that the gene predictor accounted for much of the predictive power of the IPI. The predictive model was then validated using the Lenz dataset for OS. The predictor models for CHOP and R-CHOP-treated patients effectively dichotomized patients into prognostic subgroups (CHOP: p<0.0001; RCHOP: p=0.0014; see figure below) and this difference was maintained when the subset over age 60 was analyzed (CHOP: p=0.0008; RCHOP: p=0.017). When adjusted for the IPI, the molecular predictor developed for CHOP treated patients remained robust (OS: P=.0001; HR 2.39), while the molecular predictor for RCHOP predicted OS marginally (p=.06; HR 1.82), with shorter followup than the CHOP cohort. Conclusions: Unstained slides from FFPET stored for many years may be used to investigate gene expression in lymphoma biopsy specimens for which there is mature followup. Gene risk scores based on the expression of a limited number of genes are powerful predictors of clinical outcome. Disclosures: Pollock: HTG, Inc.: Employment, Equity Ownership. Botros:HTG, Inc.: Employment, Equity Ownership. Horning:Genentech: Employment, Equity Ownership.

Abstract Abstract 429 BACKGROUND: While standard clinical prognostic factors such as those included in the International Prognostic Score (IPS) predict outcome in cHL patients, predicting the outcome of patients might be further refined using biological factors. We therefore tested whether pretreatment serum cytokines, that measure the biological activity of the tumor as well as the host response, could provide additional prognostic information in cHL patients. METHODS: Newly diagnosed cHL patients were prospectively enrolled in the University of Iowa/Mayo Clinic SPORE Molecular Epidemiology Resource (MER). Serum cytokines were measured from pretreatment blood draws using a multiplex ELISA (Invitrogen, Camarillo, CA). Thirty cytokines, including pro-inflammatory, Th1 and Th2 associated cytokines, were analyzed using the Luminex-100 system Version 1.7 (Luminex, Austin, TX). Data analysis was performed using the MasterPlex QT 1.0 system (MiraiBio, Alameda, CA). The assay was also performed on research serum draws from 50 non-lymphoma epidemiology controls enrolled on a corresponding lymphoma etiology case-control study. The upper limit of normal for each cytokine was defined as the 95th percentile from the control distribution. Clinical data were abstracted using a standard protocol and all patients were followed systematically for event-free (progression, retreatment, or death due to any cause) and overall survival (EFS and OS, respectively). Cox proportional hazards models were used to evaluate the association between cytokines and outcome. RESULTS: 140 HL patients with cHL diagnosed between September 2002 and February 2008 with available pretreatment serum were included. The median age was 40 years (range 18–89) and 53% were male. Seventy-seven patients (55%) had limited stage disease and 63 (45%) had advanced disease. One hundred and twenty one patients (86%) were treated with ABVD chemotherapy, while the remainder received the Stanford V regimen (9%), BEACOPP (4%) or MOPP chemotherapy (2%). At a median follow-up of 47 months (range 0–87), 25 patients (18%) had an event and 16 patients had died (11%). The serum levels of 12 cytokines (EGF, FGFb, GCSF, HGF, IL-6, IL-8, IL-12, IL-2R, IP-10, MIG, TNFa and VEGF) were significantly higher in cHL patients than controls. When correlated with clinical factors, IL-2R, IP-10, MIG, IL-6 and HGF were associated with an increased sedimentation rate, B-symptoms, advanced stage and splenic involvement. Elevated IL-8 levels were associated with bulky disease. Elevated levels of HGF, IL-6, IL-8, IL-2R, IP-10 and MIG were all associated with a poorer EFS, however only IL-2R (p=0.002) and IL-6 (p<0.001) were independently prognostic for both EFS and OS. Furthermore, using IL-6 and IL-2R in a 2-cytokine model (see Figure), patients with elevation of both cytokines had a dramatically greater likelihood of early relapse and death than those with normal levels or elevations of only one of the cytokines (1-year EFS 40% compared to 97% and 93% respectively, p<0.0001; 2-year OS 49% compared to 98% and 95%, p<0.0001). While elevated IL-6 and IL-2R levels correlated with the IPS (p=0.002), this 2-cytokine model remained independently predictive of prognosis. In fact, when only good risk patients (IPS 0–3) were considered, the 2-cytokine model remained predictive of patients at risk for early relapse (1-year EFS 54% versus 95% and 97%, p<0.0001). CONCLUSION: Elevated pretreatment serum cytokines are associated with an increased likelihood of disease relapse and an inferior survival in patients with cHL. The pretreatment cytokine profile, particularly serum levels of IL-6 and IL-2R, may identify cHL patients at high risk for early disease relapse who may benefit from novel therapeutic approaches. Disclosures: No relevant conflicts of interest to declare.