Cervical cancer survival in a resource-limited setting-North Central Nigeria

Abstract

Background: Organized cervical cancer screening services are presently lacking in Nigeria contributing to late presentation and diagnosis of invasive cervical cancer cases (ICCs) at advanced stages in most gynecologic units in Nigeria. We evaluated outcomes of ICCs diagnosed at Jos University Teaching Hospital (JUTH) to better understand factors associated with cervical cancer survival in similar resource limited settings.

Methods: We performed a retrospective cohort study with a prospective follow up data to estimate time from diagnosis to mortality among women diagnosed with ICCs at JUTH. Women who were diagnosed with ICCs between January 2011 and May 2013 were followed up after initial evaluation at JUTH and subsequent referral for specialized treatment in one of the national oncology treatment centers in Nigeria. The main outcome measured was all-cause mortality rate and overall survival (OS) after diagnosis of ICC. The follow up data were updated and observations were censored March 31, 2015. The overall death rate was estimated using the total number of death events and the cumulative follow-up time from diagnosis to death. We conducted Cox proportional hazard regression to assess factors associated with death.

Results: A total of 65 histologically confirmed ICCs were followed up. The median age of the cohort was 50 years with a median parity of 7. The HIV prevalence in the cohort was 8.2 % and the majority (72.3 %) were diagnosed at advanced stages (AD) of ICC. Simple total abdominal hysterectomy (TAH) was performed in 38.9 % of patients who were diagnosed at early stage disease (ED). After a cumulative follow up of 526.17 months, 35 deaths occurred with an overall death rate of 79.8 per 100 women-years. We also found a significantly higher hazard of death in women with AD (HR = 3.3) and baseline anemia (HR = 3.0). In the subgroup of women with ED, the OS was significantly higher for those who had TAH compared to those who did not (26.5 versus 11.6 months respectively).

Conclusion: Advanced stage disease and baseline anemia were independently associated with higher death rate. Cervical cancer patients diagnosed at early stages by non-oncologic specialist in settings lacking the standard of care may benefit from improve survival with simple hysterectomy.

Keywords: Advanced disease; Anemia; Cervical cancer; North-Central Nigeria; Survival.

Fig. 1

Bar chart of stages of cervical cancer at diagnosis in Jos Nigeria

Fig. 2

Kaplan Meier graph for survival probabilities of cervical cancer stages at diagnosis in Jos Nigeria. Early cervical cancer is represented with “staging = 0” and Advanced cervical cancer is represented with “staging = 1”. Log-rank p-value 0.001

Fig. 3

Kaplan Meier graph for survival probabilities of cervical cancer by baseline anemia status in Jos Nigeria. Cervical cancer patients with baseline anemia (PCV <30 %) at diagnosis is represented with “Anemia = 1” and patients who had normal hemogram (PCV ≥30 %) is represented with “Anemia = 0”. Log-Rank p-value 0.0002

Fig. 4

Kaplan Meier graph of observed and predicted survival probabilities for cervical cancer in Jos Nigeria. Early stage cervical cancer is represented with “Staging = 0” and Advanced stage cervical cancer is represented with “staging = 1”