Data analysis was performed using R software version 4.4.0, and results were written up using Microsoft Office 2010.

We determined the frequencies of viral marker testing in patients prior to initiation of chemotherapy.

We calculated the frequencies of HBs antigen and HBc antibody testing in all patients. For all HBs antigen-negative patients, we also determined the frequency of those who had tested for the anti-HBs antibody.

To compare the characteristics of patients who were screened with those who were not, t-test was used for continuous values and chi-squared test for categorical variables.

Multivariate analysis was used to determine whether or not age, gender, ALT value and cancer stage were associated with completion of hepatitis B screening. The significance threshold was P < 0.05.

A total of 312 cancer patients who received systemic anticancer chemotherapy between January 1, 2021 and December 31, 2023 in the medical oncology department of Brazzaville university hospital were included in our study. The mean age of these patients at the time of chemotherapy initiation was 54.9 ± 13.1 years, and 217 (69.6%) of them were female.

Breast cancer (141 patients; 45.2%), cervix cancer (49 patients; 15.7%) and prostate cancer (37 patients; 11.9%) were the most frequent primary cancers.

The mean ALT value was 21.7 ± 14.6 IU/mL, and 284 (91%) patients had normal ALT levels.

The basic characteristics of the study population are summarized in Table 1 .

The overall rate of HBV screening before the start of chemotherapy was 66.7% (n = 208). Only HBsAg was performed for hepatitis B screening (see Table 2 ).

The characteristics of patients who had been screened and those who had not are reported in Table 3 . Of the patients screened for HBsAg, 17 (8.2%) were positive. All patients with positive screening test were prescribed Tenofovir, an antiviral treatment.

*ULN: upper limit of the normal.

**Of the 208 patients tested for HBsAg, 191 were negative. Among these 191 patients, we looked for those who had been tested for HBsAb.

In multivariate analysis, age over 60 was significantly associated with HBV screening. No association was found for gender, ALAT value or cancer stage (see Table 4 ).

The mean age of patients in our study was 54.9 ± 13.1 years.

An average age in the fifties at the time of chemotherapy has been reported by other authors. In a study carried out in Türkiye, for example, the mean age of patients was 58 years [4] . Similarly, mean ages of 55 and 58.4 years were reported in studies conducted in Thailand and the Taiwan region, respectively [5] [6] .

Our patients were predominantly female (69.6%). In contrast to our work, a male predominance has been found in several studies [4] - [6] .

The female predominance in our work could be explained by the high awareness of female cancers compared with male cancers in our country. This could result in the diagnosis of a greater number of cancers in women than in men [7] . It has also been reported that women’s specific health events (pregnancy, menstrual problems) make them more likely than men to consult a doctor and be screened for cancer [8] .

Breast cancer was most frequent in our series (45.2%), followed by cancer of the cervix (15.7%). Like us, an author in Lebanon found a predominance of breast cancer in 34.2% of patients [9] . In a Thai and a Turkish study, the most frequent cancers were head and neck cancers and gastrointestinal cancers, respectively [4] [5] .

Regional differences in exposure to environmental risk factors, and the predominance of genetic factors for cancer in certain populations may be one explanation for these differences.

The frequency of patients screened was 66.7%.

HBs antigen was the only HBV serological marker prescribed for screening patients in our series.

The reasons why only HBs antigen was used for hepatitis B screening are not known, and further work would be required to determine them. We believe, however, that insufficient awareness of hepatitis B reactivation among those involved in prescribing chemotherapy and the lack of national or hospital guidelines on hepatitis B screening in patients eligible for cancer chemotherapy are possible explanations.

Indeed, according to Mongo’s work with physicians from 11 French-speaking African countries attending the 7th Euro-African Congress of Oncology in 2016, more than half of them (56.6%) were unaware of the risk of HBV reactivation during chemotherapy. Eighty-three people had taken part in the survey, including 68 oncologists (81.9%), and only 66.6% of those surveyed had claimed to prescribe HBV testing for patients receiving chemotherapy [10] .

Similarly, low rates of HBV screening by oncologists were reported in two studies, in the Netherlands and India. Of 110 oncologists surveyed in the Netherlands, only 12% screened all their patients for HBV, and 18% did not screen any of their patients [11] . In the Indian study, out of 161 oncologists, 19% did not prescribe any HBV screening to their patients [12] .

In studies of cancer patients, rates of hepatitis B screening prior to initiation of chemotherapy varied widely from country to country. In Japan, YAZAKI reported an overall screening rate of 70.6% [13] . Of the patients screened in his series, 48.8% were screened for HBs antigen only, and 41.3% were screened for both HBs antigen and HBc antibody. In Thailand, the rate of HBs antigen testing prior to chemotherapy was 27.7%, and the rate of HBc antibody testing was 4.6% [5] .

In the Taiwan region, the frequencies of different HBV serologies were 64.6% for HBsAg alone, 22.1% for HBsAg and HBcAb, and 0.7% for HBsAg, HBcAb and HBsAb [6] .

In a major US hospital network, including a large university hospital, community teaching hospitals and community oncology clinics, Kwak et al found 17.1% of patients who were screened for either HBsAg or HBcAb before chemotherapy [14] .

At the MAYO Clinic (USA), a similar screening rate of 16% was reported by Wi et al. [15] .

In a study of 1347 patients in Montreal, 600 (44.5%) were screened before the start of chemotherapy. Of these patients, 179 had screening request on chemotherapy orders and 421 were screened on demand from medical team [16] .

In Türkiye, the rate of HBV screening prior to chemotherapy reported in the SAMI FIDAN study was 63%. HBcAb was tested in only 18.9% of patients [4] .

It is possible that screening practices in a country are influenced by the country’s level of endemicity. Thus, countries with high endemicity would tend to screen less than countries with low endemicity. However, the low screening rate observed in Thailand, despite its high endemicity, suggests the involvement of other factors.

Differences in the recommendations of learned societies on the criteria defining patients to be screened may also be at the root of the low rates reported in the aforementioned studies.

In the year 2008, the US Centers for Disease Control and Prevention (CDC) recommended viral hepatitis B screening for all patients undergoing chemotherapy, while the 2010 guidelines of the American Society of Clinical Oncology recommended screening only for those at high risk of reactivation [17] [18] .

However, in its 2020 update, ASCO recommended, as did the American Association for the Study of Liver Diseases (AASLD), the Asia Pacific Association for the Study of the Liver (APASL) and the European Association for the Study of the Liver (EASL), that all patients planning systemic anticancer therapy should be tested for HBV [19] - [22] .

In our series, after multivariate analysis, the factor associated with the performance of HBV screening was age greater than or equal to 60 years (60 to 69 years: OR = 4.17, P = 0.002; Age ≥ 70 years: OR = 4.56, P = 0.012). The higher frequency of comorbidities in subjects over 60 may motivate a more cautious attitude prior to chemotherapy, hence a broader pre-therapeutic workup including HBV screening.

In the United States, Kwak found that the presence of hematological cancer and black race were associated with greater HBV screening [14] . A positive association between the presence of hematological cancer and HBV screening was also found by Yazaki in Japan [13] . In our case, none of the patients had hematological cancer, as only patients with solid organ malignancies are cared for in the oncology department of our center. Those with haematological cancers are cared for in the haematology department.

HBsAg was the only screening test used in our study. It was positive in 8.2% of patients. A result close to our own was reported by Ratchapong in Thailand, which like Congo is in an area of high HBV endemicity [5] .

Lower frequencies of HBs antigen-positive patients have been reported by authors in the USA and Türkiye, respectively, countries of low and medium endemicity. In the USA, the frequency of HBs antigen-positive patients was 0.9%, and in Türkiye, 2.9% [4] [14] .

HBs antigen should not be the only test to be carried out for screening prior to the initiation of chemotherapy. Some HBs antigen-negative patients may be carriers of occult hepatitis B, and therefore present a significant risk of reactivation as a result of chemotherapy. That is particularly true for patients living in Africa, where the prevalence of occult hepatitis has been estimated at 14.8% [23] .

This was a retrospective study based on patients’ medical records. Some potentially relevant data not reported in the medical records could not be taken into account. This is the case for patients’ socioeconomic status. This could have made it possible to determine whether the absence of HBV screening was due to a lack of prescription by the physician or to the patient’s inability to pay for screening.

The notion of HBV vaccination was also not regularly reported, and some patients did not know whether or not they had been vaccinated against HBV. The absence of a record of patients’ hepatitis B vaccination status in medical records made it impossible to determine the number of patients who had not been screened because they were regularly vaccinated.

Finally, our sample was relatively small. As a result, the small number of patients for certain variables could bias the search for the existence of an association between these variables and the performance of the screening test.