The development of an international childhood cancer hospital register database in 13 African countries. A project of the French African Pediatric Oncology Group (GFAOP)

Abstract

Background and aim: The establishment of an international hospital‐based register (HBR) for the French African Pediatric Oncology Group (GFAOP) was a necessary step in the group's clinical research program. With help from the Sanofi Espoir Foundation's “My Child Matters” program, the GFAOP resolved to develop an international HBR network to collect quality data on children attending the Pediatric Oncology Units (POUs). Methods: All children entering POUs from January 2016 to December 2018 were registered using an online questionnaire. Data collection included information on diagnosis, disease stage, demographics, socioeconomic status, and outcome. An intensive training program was developed to improve both data quality and quantity. Results: Among the 3348 children registered, 3230 had a suspected cancer, 681 were not confirmed. A diagnosis was confirmed on radiological, clinical, or histological examination for 2549 children including Burkitt lymphoma (516: 20%)—the most frequent diagnosis, Wilms’ tumor (459: 18%), retinoblastoma (357: 14%), and acute lymphoblastic leukemia (345: 13%). Of these, 2187 children were treated. Early deaths, abandonment, economic difficulties, and lack of equipment were some of the reasons offered to explain the numbers of undiagnosed and untreated children. Vital status is known for 1994 children: 1187 died and 807 were alive, 551 of these with a follow‐up > 12 months. Conclusion: This work has provided reliable data on children attending the POUs, especially clarifying reasons and occasions for care rupture. The data will help to identify material, human resources, and staff training needs, to evaluate progress, and to encourage consideration of pediatric cancer in national cancer plans.

The feasibility of implementing Toronto childhood cancer stage guidelines and estimating the impact on outcome for childhood cancers in seven pediatric oncology units in sub‐Saharan Africa. A study from the Franco‐African Pediatric Oncology Group

“…Since 2019, the required follow-up period was extended to 36 months. 11 Follow-up of cancer patients is a recognized challenge in LMIC. For example, 21% of the 483 patients registered were excluded from analysis in a previous study with 36-month follow up, 9 while in this study data were missing for 13% of cases at 12-month follow-up.…”

“…Each of these selected POUs registered at least 45 cases per year. The staff of each of the participating POUs included at least one pediatric oncologist, a pathologist, a pediatric surgeon, and encompassed also hematology, radiology, and ophthalmology departments located in the hospital campus or within walking distance of the POUs 11 …”

“…Training was mandatory and incorporated a seven‐module e‐learning program, including lists of diagnostic documents necessary for each cancer type (see Supporting Information on training program). Single case entry time was calculated at 30 minutes, and POUs received financial compensation of 100€ per month from the GFAOP 11,12 . Data collection and processing were quality controlled, and each POU was subjected to at least one external audit by GFAOP research coordinators and oncology experts.…”

“…The registry collects, along with other variables, information on stage of the registered cases, and therefore represent a valuable source of stage data for the few existing PBCRs in the participating countries. [10][11][12] Here, we describe the feasibility of the implementation of TG to the cancers diagnosed, staged, and registered by the seven selected POUs of the GFAOP, and estimate survival by stage at diagnosis at 3, 6, and 12 months. We compare the cancer type frequency across selected hospital units.…”

“…Data are collected prospectively online, at source in the POUs by trained researchers, and centralized in the database of the GFAOP HBCR, held in Gustave Roussy in France. The registry collects, along with other variables, information on stage of the registered cases, and therefore represent a valuable source of stage data for the few existing PBCRs in the participating countries 10–12 …”