Changing cancer incidence in Kampala, Uganda, 1991–2006

Abstract

Incidence rates of different cancers have been calculated for the population of Kyadondo County (Kampala, Uganda) for a 16-year period (1991–2006). This period coincides with continuing social and lifestyle changes and the peak and subsequent wane of the epidemic of HIV-AIDS. There has been an overall increase in the risk of cancer during the period in both sexes, with the incidence rates of cancers of the breast and prostate showing particularly marked increases (4.5% annually). Prostate cancer is now the most common cancer in men. The incidence of cancer of the esophagus, formerly the most common cancer in men and second in frequency in women, has remained relatively constant, whereas the incidence of cancer of the cervix, the most common malignancy in women, continues to increase. Since the early 1990s the incidence of Kaposi sarcoma (KS) in men has declined, and while remaining relatively constant in women, it has been diagnosed at progressively later ages. The rates of pediatric KS have declined by about 1/3rd. The incidence of squamous cell cancers of the conjunctiva has also declined since the mid 1990s. Cancer control in Uganda, as elsewhere in sub-Saharan Africa, involves meeting the challenge of emerging cancers associated with westernization of lifestyles (large bowel, breast and prostate); although the incidence of cancers associated with poverty and infection (liver, cervix, esophagus) shows little decline, the residual burden of the AIDS-associated cancers remains a major burden.

Kampala Cancer Registry (KCR) was established in 1954 with the aim of obtaining information on cancer occurrence in the population of Kyadondo County in which the capital city of Kampala is situated.1 The registry, which is located in the Department of Pathology of the Makerere University College of Health Sciences, achieved adequate coverage of the population in 1960, before activities were halted during the period of dictatorship and civil war in the 1970s and 80s. Registration of the Kyadondo population was restarted in 1989 and has continued since then. This long period of operation of KCR provides a unique opportunity to study temporal trends in cancer patterns in an African setting, because in much of Africa, there have been marked social changes in the population in last 50 years. Progressive urbanization of the population has meant that Kampala city has grown almost 10-fold, from a population of 47,000 in 1959 to 774,000 in 1991 and 1.21 million in 2002.2 Uganda was one of the countries most affected by the HIV epidemic in the late 1980s, but the prevalence of infection fell steeply between 1992 and 2002 and has now stagnated between 6.1% and 6.5% (Uganda AIDS Commission, 2008). These changes in HIV prevalence may be reflected in the trends of AIDS-related cancers.

In a previous paper3 we presented cancer incidence data from KCR for 4 time periods (1960–66; 1967–71; 1991–94; 1995–97), spanning 38 years in total, to illustrate the profound changes in cancer patterns resulting from marked social and lifestyle changes and the emergence of the epidemic of AIDS. Here we look at more recent changes, in the decades of the 1990s and 2000s, to identify whether further changes in cancer patterns are evident.

Material and Methods

KCR collects information on cancer cases resident in Kyadondo County from several sources, as well as scrutinizing records of the Department of Pathology (the host institution) tumor registrars by conducting regular search for cancer cases admitted to, or treated in the 4 main hospitals in Kampala (including the Uganda Cancer Institute and Department of Radiotherapy within the Mulago Hospital, the main teaching and national referral hospital) and also the Hospice Uganda. Copies of all reports mentioning cancer are received from 2 private pathology laboratories that deal with specimens from patients treated by private practitioners and clinics.

Data are abstracted onto notification forms and then entered into the registry database, using the CANREG system (http://www.iacr.com.fr/) which, at the stage of data entry, prevents the use of nonexistent codes and performs checks for internal consistency between variables. It also identifies potential duplicate registrations.

Tumor site and morphology were coded according to the 2nd edition of the International Classification of Diseases for Oncology (ICD-O).4, 5 For tabulation of results, these were converted to the 10th revision of the ICD.6 Squamous cell carcinomas (SCCs) of the conjunctiva were defined as tumors with ICD-O morphology codes M8010–M8082 of conjunctiva (ICD-O C69.0) or eye, unspecified (ICD-O C69.9); only 3% of SCCs occur at other subsites of the eye.

Population

Population censuses were performed in 1991 and 2002, and for these years, the population of Kyadondo County was available by sex and 5-year age group (Fig. 1). Intercensal estimates were prepared assuming a constant rate of change, within age–sex groups. The Department of Statistics of the Ministry of Finance and Economic Planning estimates that, since 2002, the population has been growing at the rate of 3.37% annually; this rate of change was assumed to apply to all age sex groups, to prepare estimates of the population in the period 2003–2006.

Figure 1. Population of Kyadondo County, by age and sex, censuses of 1991 and 2002.

Statistical methods

Incidence rates were calculated for 5-year age groups by sex, for each year (1991–2006) and for 3 time periods: 1991–1995, 1996–2001, 2002–2006. Age-standardized rates (ASRs) were calculated using the world standard population.7 Average annual percentage change (aapc) (and 95% confidence intervals) over the whole 16-year period was calculated for each sex, for those cancer sites with at least 1 case registered in each year, assuming a constant rate of increase or decrease in the period considered. Graphs of time trends in rates use 3-year moving average values of rates to minimize fluctuation due to small numbers of cases.

Results

During the 16 years of registration considered (1991–2006) a total of 16,093 cases (7,267 male and 8,826 female) were registered. Figure 2 shows the trends in incidence rates (3-year moving average ASRs) in males and Figure 3 in females. Incidence rates have increased in both sexes over the period; the aapc was +1.3% (95% CI: 0.0–2.7) in males and +3.2% (95% CI: 2.4–4.1) in females. The ASR for the period 1991–1995 was 146.2 per 10⁵ in males and 145.3 per 10⁵ in females, and in 2002–2006 it was 167.9 per 10⁵ in males and 206.9 per 10⁵ in females (Table 1).

Figure 2. Trends in age-standardized incidence rates of selected cancers in males, 1991–2006 (3-year moving averages).

Figure 3. Trends in age-standardized incidence rates of selected cancers in females, 1991–2006 (3-year moving averages).

Table 1. Age-standardized incidence rates (per 100,000) in 3 time periods, and average annual percentage change in age-standardized rates in the period 1991–2006

In males, the most common cancers over the period (in terms of ASRs) are cancers of the prostate and Kaposi sarcoma (KS; Table 1); they show opposite trends in rates, with significant increases in the incidence of prostate cancer (4.5% annually; 95% CI: 0.9–8.0) and decreases in the incidence of KS (−2.8% annually; 95% CI: −4.6, −1.0). The incidence of cancer of the esophagus has remained relatively constant (Fig. 2). Among the other sites with lower rates of incidence, increases are evident for non-Hodgkin lymphomas and lung cancer (the latter nonsignificant), whereas rates of large bowel and liver cancers are more or less constant. There has been a slight, nonsignificant decline in incidence of penile cancer (aapc of −1.4%; 95% CI: −4.8, 2.0).

In females, the most frequent cancer over the whole period was cancer of the cervix uteri. The rates have been increasing over most of the period (aapc 3.0%; 95% CI: 0.9–5.1) so that the ASR in 2002–2006 was 52.4 per 10⁵ (Table 1). The incidence of breast cancer is about half that of cervix cancer, but increases have been even more marked: +4.5% (95% CI: 2.6–6.5) annually since 1991. For both cancer of the cervix and breast, the increases in incidence rates are more evident at older age groups (after age 50) than among the young (Fig. 4). In contrast to males, the rates of KS in females have been more or less constant.

Figure 4. Age-specific incidence rates of cancers of the cervix and breast in the periods 1991–1995, 1996–2001 and 2002–2006.

Among the other sites with lower rates of incidence, significant increases in rates are evident for non-Hodgkin lymphomas and stomach, large bowel and lung cancers, whereas rates of esophageal cancer have shown marked fluctuations over the period, with no obvious overall trend.

Figure 5 shows the age-specific incidence rates of KS in 3 time periods, in males and females. In both sexes, there has been a decline in incidence of KS in the youngest age group 0–4. In males, the decrease in incidence has concerned particularly the high rates noted in the early 1990s at ages 30–44. In females, in contrast, the relatively constant incidence has been accompanied by a shift in age distribution, from a peak rate at ages 25–29 in 1991–1995, to 35–39 in the later time periods. The changes in the mean age at diagnosis are shown in Figure 6. From being significantly younger than men with KS, the mean age of female cases increased in the late 1990s and there is now no significant difference in the mean age at diagnosis between the sexes.

Figure 5. Age-specific incidence rates of Kaposi sarcoma in males and females in the periods 1991–1995, 1996–2001 and 2002–2006.

Figure 6. Mean age at diagnosis (with 95% confidence limits) of Kaposi sarcoma, by year, in males and females.

Figure 7 shows the trends in incidence (3-year moving average) of conjunctival squamous cell cancers (both sexes); the incidence rates increased to a maximum in 1996 and then declined.

Figure 7. Age-standardized incidence rate per 100,000 of squamous cell conjunctival cancers, by year (3-year moving average).

In the childhood age range 0–14, 1,515 cancers were registered in the 16-year period, the most common being non-Hodgkin lymphomas (33.1%), KS (27.3%), Wilms' tumors (6.3%), leukemias (5.9%) and retinoblastomas (5.1%). Within the non-Hodgkin lymphoma group, more than 1/5th were of unspecified type and over 60% were Burkitt lymphomas (20.3% of all childhood cancers). The profile of childhood cancer changed over the period. KS was the most frequently diagnosed in 1991–1995 (ASR, 62.6 per million), but the incidence had fallen to 44.6 per 10⁶ by 2002–2006. Conversely, the incidence rates of non-Hodgkin lymphoma rose from 42.2 per 10⁶ in 1991–1995 to 67.9 per 10⁶ in 2002–2006, with 2/3rd specified as Burkitt lymphomas. The mean age of children with Burkitt lymphoma remained constant (7.0–7.3 years).

Discussion

To study time trends, it is important that the degree of completeness of registration of incident cancer cases should be similar throughout the period under consideration. In recent years (since 1991) we believe that registration has been relatively complete. Evaluation by independent case ascertainment for the period 1994–1996 suggested that the registry had identified 90% of incident cancers.8 The results from the registry have been published in Cancer Incidence in Five Continents, volumes VII (1991–1993), VIII (1993–1997) and IX (1998–2002).9–11

In addition to completeness of ascertainment, a valid estimation of incidence rates requires that accurate population denominators are available. For Kyadondo County, census counts (by age group and sex) of the population were available for 1991 and 2002, and we made use of interpolations and projections for the other years. There must be some question as to the accuracy of the person-years at risk estimates, particularly with respect to the age distributions, although no better data are, of course, available.

In common with much of urban Africa, lifestyles in Kampala are changing rapidly, as the population changes from one comprising relatively recent immigrants from village life, to one of 2nd-generation inhabitants, engaged in wage-earning or the informal economy, and purchasing foodstuffs and other necessities, rather than producing themselves. This demographic transition is accompanied by familiar trends in patterns of health and illness, with a decrease in maternal and infant mortality and a rise in the importance on noncommunicable diseases.12 It is therefore not surprising to note a steady increase in the incidence of cancer in both sexes, so that, compared to the average for “developed” countries,13 overall cancer incidence in females in Kampala is now similar, although it is still little more than half in males. One might reasonably expect the changes to relate to an increase in the cancers that are particularly common in “western” populations (Europe and North America; lung, prostate, large bowel and breast) and a decrease in those more familiar in the so-called developing world (cancers of the stomach, esophagus, liver and cervix uteri).

Cancer of the prostate became the most common cancer in men (in terms of age-standardized incidence) in 1996 and there has been a major increase in incidence 4.5% annually on average over the 16-year period. This cancer was relatively rare in the 1960s (ASRs of 3–6 per 10⁵),3 but by 2002–2006 the age-standardized incidence rate (39.6 per 10⁵) was one of the highest observed in Africa.13 Most of this increase is in elderly men aged 65 or over. The increase in Uganda is certainly not due to screening, although it is quite likely that increased awareness, a greater readiness to perform prostatectomy for urinary symptoms in elderly men and histological examination of operative biopsies have played a role.

The incidence of breast cancer in women has been increasing at the same rate (4.5% annually), although the absolute incidence rate (ASR of 31.0 per 10⁵ in 2002–2006) remains low by world standards.13 Most of the change has been in postmenopausal women (Fig. 4). It is possible that some of the increase is related to declines in fertility, or increases in levels of overweight/obesity. Although there are no data on the extent of such changes in the Kampala population, census data suggest that urban dwellers, and those with higher educational levels, have lower than average fertility, but it is unlikely that changes in these parameters could explain such large increases in incidence rates.

Tumors of the large bowel are rather rare (as noted by Burkitt14), although there are increases occurring, especially among women. Likewise, cancers of the lung remain relatively rare although incidence is increasing in both sexes; this may relate to increases in smoking prevalence. According to WHO,15 16.3% of adult males and 2.2% of females were daily tobacco users in 2001–2006, although there are no long-term trend data available.16

The increase in incidence of cancers associated with a “western” lifestyle (breast, prostate, large bowl, lung) is not, however, being accompanied by major declines in the cancers traditionally associated with East Africa. Cancer of the esophagus was the most common cancer in men in the 1950s and 1960s. The reasons are not well understood. Although tobacco and alcohol certainly play a role, other factors have been intensively investigated as responsible for the relatively high rates in East and South-East Africa, especially dietary deficiencies and mycotoxins.17, 18 However, esophageal cancer rates appear to have changed little in the last 16 years (constant in males, fluctuating around the same level in females) when one might have expected the diseases associated with nutritional deficiency (or contamination) to have declined.

Likewise, stomach cancer has not shown the declines in incidence rates observed in western countries—rates have in fact been increasing. This may be the result of better diagnosis because of increasing access to gastroscopy. Prevalence of infection with Helicobacter pylori in adults in Kampala was 87% in a recentstudy, and, interestingly, there was a significantly higher prevalence in younger subjects than those over 50.19

The incidence of liver cancer in Uganda is relatively low in comparison with other parts of sub-Saharan Africa, despite a relatively high prevalence of chronic carriage of hepatitis B20 and aflatoxin contamination of foodstuffs.21 The incidence of liver cancer seems to be increasing, especially among females, although there is no evidence that prevalence of infection with Hepatitis B has changed substantively; it is possible that the rates have been influenced by other factors such as diabetes or increasing obesity.22

Cancer of the cervix uteri has been the most common cancer in women ever since the inception of the registry1 and it so remains. The rates are still increasing at an average of 3% a year and are now very high (ASR 52.4 per 10⁵ in 2002–2006). The reasons for these changes are not immediately clear. They are unlikely to be related to the epidemic of AIDS, for, although cervix cancer is considered to be an AIDS-defining condition in the United States, it is not clear whether the association is simply due to the increased prevalence of infection with oncogenic human papillomaviruses (HPVs).23 In any case, most of the increase in incidence has been in older (postmenopausal) women (Fig. 4). The social disruption of 1970s and 1980s may have favored the spread of HPV (like other sexually transmitted diseases), resulting in increasing risk of cervical cancer. Antibodies to types 16, 18 or 45 were found in 17% of women in a study in Kampala,24 and there was no decline in prevalence of infection with age, as is generally observed.25 Screening is confined to a very limited number of opportunistic examinations.

In the comparison of incidence rates between the 1960s and 1990s3 the cancers showing the most marked increases in incidence were those related to infection with HIV, particularly KS, and SCC of the conjunctiva. In the 1960s, KS was of the typical “endemic” pattern, involving the skin, particularly the legs, and affecting principally males, with the risk rising progressively with age. Human herpes virus 8 (HHV8) is the etiological agent responsible for KS.26 HHV8 has been identified in over 85% of KS tissue specimens in Uganda,27 and seroprevalence studies suggest a relatively high prevalence of infection by HHV8 in the general population of Uganda—considerably higher than in the United States and Europe, which would be consistent with the elevated frequency of endemic KS which was present before the epidemic of HIV-AIDS.28, 29 The enormous increase in incidence of KS between the 1960s and the early 1990s was accompanied by a narrowing of the sex ratio (from 18:1 in 1960–1971 to 1.7:1 in the 1990s) and a marked increase in the incidence of KS in children. These changes were the result of the evolution of the epidemic of HIV-AIDS in Uganda, and the age-specific incidence of KS was noted to correspond closely to the age-specific reporting rates for AIDS.30 The prevalence of HIV infection among pregnant women had reached 30% in the urban population of Kampala in 1990–1992, but since that time it has declined to around 10.1% in 2005.31 There has been a fall in incidence of KS in males since the early 1990s, although the incidence in women has remained almost constant. One might have expected some change in the age-specific incidence of KS over time, as antiretroviral therapy becomes available, albeit to a limited number of sufferers. Although this does not seem to have occurred in males (peak age, and mean age at diagnosis of 35.0 have changed little in the period), in women, the mean age of onset seems to have increased from around 28 in 1991–1995 to 32 in 2002–2006, with the peak incidence, as in males, at 35–39 in more recent periods.

SCC of the conjunctiva has been shown to be associated with HIV infection in African populations,32, 33 and the incidence in the Kampala population increased 10-fold between 1960–1971 and 1995–1997.30 There appears to have been some decline in the incidence of these cancers in the Kampala population, coinciding with a decline in HIV prevalence.

The incidence of non-Hodgkin lymphomas has also increased markedly during the period studied. This may also be related to improved survival of patients with HIV infection, permitting a more prolonged duration of immunosuppression and the development of more clinically evident lymphomas.

Incidence rates of penile cancer, which was clearly very frequent in the early case series from Kampala,1, 34 had decreased between the 1960s and 1990s,3 and the rates continue to decline (although remaining relatively high—the ASR of 2.7 per 100,000 in 2002–2006 can be compared to 0.6 in the black population in the United States SEER Registries11). Penile cancer has been related to genital hygiene35 and the decline in incidence may be related to improved hygiene as a consequence of urbanization and greater availability of piped water supplies.

In the childhood age group, the decline in the incidence of KS may be related to improved availability of antiretrovirals and counseling of HIV-infected pregnant women, which can decrease the prevalence of congenital HIV infection in infancy.36 In contrast, the reason for the increasing incidence of non-Hodgkin lymphoma (predominantly Burkitt lymphoma) is obscure. The risk of endemic Burkitt lymphoma in childhood is not much increased by HIV infection37; perhaps an increase in the prevalence (or severity) of other risk factors—especially malarial infection in childhood—is responsible.

Conclusions

The ongoing surveillance system provided by the registry in this predominantly urban East African population provides a picture of the evolution of the problem of cancer in modern sub-Saharan Africa. The increase and subsequent waning of the epidemic of HIV-AIDS is mirrored by the trends in AIDS-related cancers, while cancers associated with westernization of lifestyles (breast and prostate cancers in particular, large bowel cancers less certainly) continue to increase. At the same time, there has been no decline, and even some increase, in the incidence rates of the major cancers of this region—cervix, liver and esophagus, although so far there is little evidence for the onset of the epidemic of tobacco-related cancer which has been such an important feature of the cancer profile in economically developed countries. However, although economic hardship may limit tobacco consumption and its health consequences, it will also restrict the capacity to mount effective programs of cancer control.

Acknowledgements

The authors acknowledge the cooperation and assistance of the medical nursing and records staff of the Mulago, Rubaga, St. Francis (Nsambya), and Mengo Hospitals, and the Uganda Hospice, for their assistance in tracing and recording cancer patients, as well as the pathologists and hematologists for giving access to their laboratory records.