Signal Detection in Monoclonal and New Agents in Lymphoma

Overview

Non Hodgkin lymphoma (NHL) is the 5th cancer in France. Advances in NHL therapy have resulted in improved cure rates with a 5 year relative survival rate estimated at 55% and a 5-year prevalence estimate of 27,750 cases. Since 2000, the addition of anti-CD20 antibody to the standard treatment regimen composed of cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) led for the first time to decline of the specific mortality. After treatment, patients with NHL experienced elevated risks for therapy-related leukemia, several solid tumors and late toxicities such as cardiovascular or neuro-psychiatric diseases which impact on quality of life. However little is known concerning long-term toxicity of this class of new agents so called "targeted drugs" such as anti-CD20. The primary objective of this cohort study is to estimate long term toxicity in NHL patients (i.e. 10 to 20 years) using data already collected (i.e. internal analysis) and to compare drugs consumption to that of controls (i.e. external analysis).

The French public healthcare system perspective developed a unique SNIIR-AM (interregimens National Information Health Insurance system.) data base that contains up to 18 TeraOctet of medical consumption data (consultations, exams and drug prescriptions and hospitalization, yes/no). Its main limitation to detect disproportionate consumption after cancer care is the lack of data available on socio-demographic, pre-diagnostic patient and disease characteristics. The LYSA (Lymphoma Study Association) is a national cooperative group which is world pioneer in NHL treatment with novel agents. Patient's bio clinical and demographic characteristics (n=10.000) are fully monitored and stocked in a data base registered in 2006 by the French National Cancer Institute (INCA). The cooperative network of 100 LYSA centre allows following all the patients on long term (up to 20 years after diagnosis). Then, the mixture of these two sources of data will offer a unique material at the nation scale to detect signal of late onset disease, possibly linked to the initial cancer treatment. Statistical Methods Disproportionate reporting of drugs consumption in fields of neuro-psychiatric troubles, cardio-vascular diseases, anti-cancer agents, antibiotics and antiviral agents will be extract from the French public healthcare database (external analysis). An economical study is embedded. As the LYSA data contained all fully monitored patients' bio clinical and demographic characteristics, special attention will be paid to the addition of new agents, in particular anti-CD20 antibody, to the cytotoxic chemotherapy regimens in conjunction with other well-known risk factors (e.g. tobacco, obesity, comorbidities; internal analysis). Data will be reported according to the initial NHL treatment strategy and cross-validated by long-term medical consultations. Multivariate analysis will be computed to account for competing risks. Results No studies have yet been published in France on the use of care issues in long remission or cure of cancer. There are very few international publications on this subject and they lacked of statistical power or follow up. The present project follows the recommendation from the French Medicine Academy which in its March 15, 2011 decision call for more rigorous follow up of new agents in Oncology. In addition, this project is in concordance with the so named "plan cancer 2" from the INCA which aimed to focus on sequels in cancer survivors. If the results of this study are relevant, the investigators should have to run through the consumption of care, a good objective (unbiased) indicator of healing and Health-related quality of life (HRQL) in cancer survivors. Such an indicator could help to learn about survivors recovery after their cancer, with or without differences according to the patient characteristics, initial location of the cancer, its treatment and possible iatrogenic complications.