healthmedicinet

HBV/HCV seroprevalence

This study presents a prospective assessment for the seroprevalence of HBV and HCV
in patients receiving cytotoxic chemotherapy for solid tumors in a country with low
endemicity of HBV and HCV. Other studies examining these factors exist but the countries
assessed were high endemic 3], 6], 8] or intermediate endemic 19]–21] prevalence.

In our population, the prevalence of chronic HBV [HBsAg (+)] was lower than that of
the general population in France (0.3 % vs. 0.65 %) 1] but higher (0.3 % vs. 0.11 %) than the French hospital prevalence 22]. The target population for selective screening consists of subjects exposed to HBV.
Seroprevalence in these patients (8.5 %) was identical to the French general population
favoring the extrapolation of these results 1].

The seroprevalence of HCV (1.3 %) was higher when compared to both the French general
(0.84 %) 1] and hospital populations (0.33 %) 22]. One contributing factor for this observation may have been due to an increased risk
of receiving a transfusion or other procedural contamination prior to 1992 due to
the age of these patients.

Rationale for selective screening

Systematic HBV screening in patients with solid tumors may not be as cost-effective
as chemotherapy when the HBVR and the prevalence of HBV are low 23]. The estimated cost of care and cost-effectiveness ratio is considered relevant.
Cost effectiveness depends on the care system of the country and therefore studies
of cost -effectiveness are difficult to extrapolate from one country to another.

The HBVR risk depends on three components: the type of solid tumor, the treatment
and the serological status. This risk was more prevalent in breast cancer (estimated
up to 41 %) 6], 7] and hepatocellular carcinoma patients (36 %) 24]. In other solid tumor, HBVR risk was approximately 16 % in prospective studies 3], 7]. Our population did not include patients with breast cancer and included only one
patient with liver cancer; making the RVHB risk low.

The highest rates of HBVR in patients undergoing chemotherapy for solid tumors occurred
with anthracycline-based regimen 6], 7]. This type of regimen is rarely used outside of breast cancer (FEC/AC); as reflected
in our study with only 4 patients who received an anthracycline.

According to the serological status, 9 patients were considered at risk of reactivation
(2 %), which shows a possible benefit of screening. But the only patient treated was
HBsAg (+). Among 388 screened patients only one was treated, which probably decreases
the relative cost-effectiveness of a routine screening. Moreover, the only solid tumor
cost-efficiency study suggested a screening with HBsAg alone 23], since the HBsAg (+) patients are most at risk of HBVR.

The screening questionnaire shows the need for oncologists to be educated on risk
factors for carriage of HBV/HCV since only 33 % identified the place of birth in endemic
areas as a risk factor (while being the main risk 15], 25]).

We found that our population was ideal for a selective screening. To our knowledge,
and from the time of this publication, there have been no published evaluations for
the selective screening process considering sensitivity, specificity, and predictive
value in a low endemic country for viral hepatitis.

Relevance of targeted screening on HBV/HCV risk factors

Selective screening is performed in two steps: a pre-screening (questionnaire), followed
by the serological test. A quality pre-screening test will focus on sensitivity while
a quality serological test will focus on specificity.

In the present study, the sensitivity for HBV was 45.5 %, leaving out more than half
of seropositive HBV patients. Additionally, the only patient HBsAg (+) would not have
been identified and treated using a selective strategy since he had no risk factor
on the questionnaire. The overall sensitivity (exposed to HBV or HCV) was 50 %, therefore
insufficient. Sensitivity was 100 % for HCV but, as stated previously, HCV infection
is not the main concern. The specificity of the questionnaire was also insufficient
(56 %, approximately).

The probability of being seropositive for HBV, HCV and one or the other in case of
positive questionnaire was very low since the overall PPV did not exceed 11 %. Recalling
the differences in clinical outcome of RHCV, the screening recommendations for HBV
and HCV cannot be the same.

Few studies have assessed the relevance of targeted screening on HBV / HCV risk factors
26]–28]. One study with pregnant women in the United States was designed to test a questionnaire
recommended by the ACIP (Advisory Committee on Immunization Practices). This questionnaire
was used for 692 parturient women among whom 8.5 % were HBV positive. The sensitivity
was less than 60 % for screening carriers of HBV28] . These findings led to the current policy of universal prenatal screening.

Limitations

This study is missing data

It was noted that serologic test were not ordered due to either (i) deliberate from
relevance deemed insufficient by the oncologist because of the entry into advanced
palliative/terminal phase or (ii) unintentional due to lack of awareness of oncologists
despite the study implementation. Without a doubt, there is an underestimation for
the risk of RHBV by oncologists 15] although it is possible to conclude that digestive oncologists may be more informed
on this since patients with gastrointestinal tumors were more likely to be tested
in our study.

This questionnaire is lacking in sensitivity. First, the main mode (one third of cases)
of transmission of HBV,1] is sex. Except for HIV seropositivity and oting relatives infected with viral hepatitis, sexually transmitted infections were intentionally excluded from the questionnaire due to the subjectivity of the
concept of unsafe sex and the lack of reliability of predictable responses. Secondly,
lack of sensitivity of our survey for HBV could be expected; actually, in one third
of cases the mode of transmission of hepatitis B is not found 1]. However, it had to be proven; we can conclude that, in our population, if a screening
must be done, it has to be systematic, especially since serologic test is sensitive,
specific and inexpensive.

The CDC (Centers for Disease Control and Prevention) developed a on-line questionnaire
specifically for these risk factors which was completed by patients 29]. The estimated time to complete this questionnaire was 5 minutes 29]. This questionnaire is being used in an ongoing prospective study at the MD Anderson
Cancer Center (USA), which is to include 3,400 patients prior to chemotherapy and
to compare both strategies in terms of sensitivity, specificity and cost-effectiveness
30]. This questionnaire could have a different sensitivity. The reliability of patients’
anamnestic data and their lack of education for risk of contamination (i.e. infected
patients who do not recognize or report on risk factors) remain to be an issue.