We conducted a cross-sectional descriptive study using data from consecutive patients enrolled in the GENCAF program between May 2022 and December 2023. The program was implemented across three major oncology centers in Yaounde: Yaounde General Hospital, Yaounde Central Hospital, and the Chemotherapy Solidarity Non-Governmental Organization (SOCHIMIO). Together, these institutions account for an estimated 80% of cancer consultations in the capital region.

Inclusion criteria were: 1) all cancer patients suspected to have a hereditary component based on at least one of the following clinical red flags: breast or ovarian cancer diagnosed at age ≤ 45 years; triple-negative breast cancer at any age; two or more first- or second-degree relatives with breast, ovarian, colorectal, gastric, or prostate cancer; bilateral or multicentric breast cancer; male breast cancer; or personal/family history suggestive of a recognized hereditary syndrome (Lynch syndrome, Li-Fraumeni syndrome, familial adenomatous polyposis); 2) completion of pre-test genetic counseling; and 3) successful saliva-based DNA collection and valid sequencing results. Enrollment was consecutive. Of 123 patients assessed for eligibility, 112 attended pre-test genetic counseling. Of these, 94 (83.9%) proceeded to germline multigene testing. Reasons for non-testing included refusal after counseling (n = 14, primarily citing fear of learning about the existence of mutations), financial barriers (n = 3), and inadequate saliva sample (n = 1). Patients under 18 years of age (n = 5 of those initially approached) were excluded before counseling. All 94 tested patients yielded valid sequencing results.

Pre-test genetic counseling was performed using a standardized workflow adapted from NCCN and ESMO guidelines to the Cameroonian context, covering hereditary cancer syndromes, possible testing outcomes (P/LP, VUS, negative), treatment and surveillance implications, and psychosocial considerations. Post-test counseling delivered actionable findings and cascade testing advice. Family history of cancer was defined as at least one first- or second-degree relative with a documented cancer diagnosis, regardless of cancer type. Family history was collected through structured interview and three-generation pedigree construction during pre-test counseling.

DNA was collected from saliva using Oragene OG-600 collection kits, stabilized at room temperature, and shipped to a Clinical Laboratory Improvement Amendments (CLIA)-certified international partner laboratory (Color Genomics, Burlingame, CA, USA). DNA extraction followed manufacturer instructions with fluorometric quantification (Qubit).

All samples underwent next-generation sequencing (NGS) using a 29-gene hereditary cancer panel: APC, ATM, BAP1, BARD1, BMPR1A, BRCA1, BRCA2, BRIP1, CDH1, CDK4, CDKN2A, CHEK2, EPCAM, GREM1, MITF, MLH1, MSH2, MSH6, MUTYH, PALB2, PMS2, POLD1, POLE, PTEN, RAD51C, RAD51D, SMAD4, STK11, TP53. Mean coverage was >250×; minimum per-base coverage >20× for >99% of targeted bases. Variant calling followed GATK-based workflows with alignment to GRCh38, annotated via ClinVar, gnomAD (global and African datasets), and LOVD.

All variants were classified per American College of Medical Genetics and Genomics and Association for Molecular Pathology (ACMG/AMP) 2015 guidelines [21] into P, LP, VUS, likely benign, or benign. Only P, LP, and VUS variants were reported. Copy-number variant detection was performed using depth-of-coverage algorithms.

Descriptive statistics were calculated for all variables. Fisher’s exact test assessed associations between P/LP status and family history. Age comparisons across gene groups used one-way ANOVA. p < 0.05 was considered significant. Analyses were performed using SPSS version 23.

Ethical approval: N˚2021/12/1424/CE/CNERSH/SP—National Human Health Research Ethics Committee, Cameroon. The study was conducted in accordance with the Declaration of Helsinki. Identifiable data were not used in this analysis.

A total of 94 patients underwent successful multigene germline testing. The cohort was 95.7% female (n = 90; 4 male). Mean age at diagnosis was 43.4 years (SD ± 11.5; range 19 - 71). Age distribution: <40 years 45.7% (n = 43), 40 - 49 years 22.3% (n = 21), ≥50 years 31.9% (n = 30). Table 1 summarizes cohort characteristics.

Table 1. Cohort characteristics (N = 94).

According to cancer type distribution, breast cancer was the most represented cancer type at 87.2% (Figure 1). A positive family history of cancer—defined as at least one first- or second-degree relative with a documented cancer diagnosis—was reported by 57 patients (60.6%). Among breast cancer patients, a notably high proportion presented with triple-negative disease—a subtype strongly associated with hereditary BRCA1 mutations. High-risk clinical features were common: 45.7% were diagnosed before age 45, and 60.6% reported a positive family history of cancer.

Figure 1.Cancer type distribution among GENCAF patients (N = 94).

Among 94 participants, 25 carried at least one pathogenic (P) variant (26.6%) and 1 carried a likely pathogenic (LP) variant (1.1%), for a combined P/LP yield of 26 patients (27.7%). Table 2 presents the distribution of pathogenic/likely pathogenic variants by gene amongst the mutated genes, amongst the population and the syndromes related while Figure 2 presents the distribution of P/LP variants by gene.

Table 2. Distribution of pathogenic/likely pathogenic variants by gene amongst the mutated genes and amongst the population (N = 94).

Figure 2. Distribution of P/LP variants by gene and associated syndromes (N = 94; 26 P/LP carriers, 27.7%).

BRCA1 overwhelmingly dominated the mutational landscape, accounting for 69.2% of all P/LP findings.

VUS were identified in 25 patients (26.6%), distributed across 13 distinct genes. Table 3details VUS distribution by gene, andFigure 3 details VUS distribution across 13 genes.

Table 3. Distribution of variants of uncertain significance by gene (N = 94).

Figure 3.VUS distribution across 13 genes (N = 94; 25 patients with VUS, 26.6%). Colors indicate biological pathway. VUS: variant of uncertain significance; FAP: familial adenomatous polyposis; MMR: mismatch repair.

When combining P/LP and VUS findings, 43 patients (45.7%) carried at least one variant of any classification. It is important to note that VUS findings do not constitute evidence of hereditary cancer risk: their clinical significance remains uncertain pending reclassification through functional studies, segregation analyses, or updated population databases. This combined figure is reported to characterize the genomic complexity of the cohort, not to imply pathogenicity of VUS. Fifty-one patients (54.3%) had completely negative results across all 29 genes.

Eight patients (8.5%) carried a P/LP variant in one gene and a co-occurring VUS in a different gene. Five additional patients carried two VUS in different genes. Representative combinations included BRCA1 P + PALB2 VUS (3 patients), BRCA1 P + ATM VUS (2 patients), BRCA1 P + APC VUS (2 patients), and MSH2 P + PMS2 VUS (1 patient). The recurrence of PALB2 VUS among BRCA1 P/LP carriers warrants longitudinal follow-up and functional characterization.

Among patients with a positive family history of cancer, 33% carried a P/LP variant, compared to 16% among those without family history (Fisher’s exact p = 0.12). Although the difference did not reach statistical significance, the trend supports genomic screening even in the absence of a documented family history, given the high rate of early-onset cancers in a context where hereditary cancer has been systematically undiagnosed. Figure 4 presents the pedigree of a 39-year-old female with a positive family history of cancer who presented with right breast cancer and a pathogenic mutation of PMS 2 gene.

CA:cancer.

Figure 4.Pedigree of a 39-year-old female with a positive family history of cancer who presented a right breast cancer and a pathogenic mutation of PMS 2 gene.

BRCA1 P/LP carriers were diagnosed at a mean age of 37.8 years (n = 17), significantly younger than non-carriers (mean 45.1 years; t-test p = 0.007). The age analysis includes 17 of the 18 BRCA1 P/LP carriers: the one likely pathogenic (LP) carrier (C0066) was excluded from the inferential age comparison due to the lower certainty of her classification, though her age at diagnosis (57 years) is reported descriptively. Age data were available for all 94 participants. Gene groups with n = 1 - 2 (BRCA2, PALB2, ATM, MSH2) are reported descriptively and were not included in inferential group comparisons. BRCA2 carriers averaged 52.0 years (n = 2), PALB2 carriers 56.0 years (n = 2), ATM carriers 56.0 years (n = 1), and MSH2 carriers 48.0 years (n = 1). These figures are reported descriptively given the small group sizes (Figure 5).

Figure 5.Mean age at diagnosis by gene group.

Several limitations warrant consideration. First, single-city recruitment at tertiary centers in Yaounde limits generalizability to rural and regional populations. Second, breast cancer is overrepresented (87.2%), limiting power for inferential analyses across other cancer types. Third, the absence of matched tumor sequencing prevents direct comparison of somatic and germline findings. Fourth, statistical power for subgroup analyses is limited by the sample size (N = 94). Fifth, functional validation of novel VUS was not performed. Despite these limitations, this dataset represents a foundational reference for hereditary cancer genomics in Central Africa.