No abstract

ObjectiveTo examine 30-year time trends in incidence, survival and mortality of lymphomas by subtype in Manitoba, Canada.MethodsLymphoma cases diagnosed between 1984 and 2013 were classified according to the 2008 WHO classification system for lymphoid neoplasms. Death data (1984–2014) were obtained from the Manitoba Vital Statistics Agency. To examine time trends in incidence and mortality, we used joinpoint regression to estimate annual percentage change and average annual percentage change. Age–period–cohort modelling was conducted to measure the effects of age, period and cohort on incidence and mortality time trends. We estimated age-specific and standardised 5-year relative survival and used Poisson regression model to test time trends in relative survival.ResultsTotal Hodgkin lymphoma (HL) incidence in men and women was stable during the study period. Age-standardised total non-Hodgkin lymphoma (NHL) incidence increased by 4% annually until around 2000, and the trend varied by sex and NHL subtype. Total HL mortality continuously declined (by 2.5% annually in men and by 2.7% annually in women), while total NHL mortality increased (by 4.4% annually in men until 1998 and by 3.2% annually in women until 2001) and then declined (by 3.6% annually in men and by 2.5% annually in women). Age-standardised 5-year relative survival for HL improved from 72.6% in 1984–1993 to 85.8% in 2004–2013, and for NHL from 57.0% in 1984–1993 to 67.5% in 2004–2013. Survival improvement was also noted for NHL subtypes, although the extent varied, with the greatest improvement for follicular lymphoma (from 65.3% in 1984–1993 to 87.6% in 2004–2013).ConclusionsTime trends were generally consistent with those reported in other jurisdictions in total HL and NHL incidence, but were unique in incidence for HL and for NHL subtypes chronic/small lymphocytic leukaemia/lymphoma, diffuse large B cell lymphoma and follicular lymphoma. Survival improvements and mortality reductions were seen for HL and NHL in both sexes.

Purpose: We performed detailed genomic analysis on 87 cases of de novo diffuse large B-cell lymphoma of germinal center type (GCB DLBCL) to identify characteristics that are associated with survival in those treated with R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone). Experimental Design: The cases were extensively characterized by combining the results of IHC, cell-of-origin gene expression profiling (GEP; NanoString), double-hit GEP (DLBCL90), FISH cytogenetic analysis for double/triple-hit lymphoma, copy-number analysis, and targeted deep sequencing using a custom mutation panel of 334 genes. Results: We identified four distinct biologic subgroups with different survivals, and with similarities to the genomic classifications from two large retrospective studies of DLBCL. Patients with the double-hit signature, but no abnormalities of TP53, and those lacking EZH2 mutation and/or BCL2 translocation, had an excellent prognosis. However, patients with an EZB-like profile had an intermediate prognosis, whereas those with TP53 inactivation combined with the double-hit signature had an extremely poor prognosis. This latter finding was validated using two independent cohorts. Conclusions: We propose a practical schema to use genomic variables to risk-stratify patients with GCB DLBCL. This schema provides a promising new approach to identify high-risk patients for new and innovative therapies.

Clinic (outside the submitted work). Dr Montori is the chair of the board of The Patient Revolution (patientrevolution.org). He receives no income from his participation in this nonprofit organization focused on advancing patient-centered care nor from the sales of his book, Why We Revolt, all of which go to fund The Patient Revolution. The other author reports no competing interests.