Chronic myeloid leukaemia in South Africa

Abstract

Despite the challenges of a resource-limited environment, the outcome of chronic myeloid leukaemia (CML) patients in South Africa is similar to that in developed countries, thanks to access to tyrosine kinase inhibitors through patient assistance programmes and clinical trials. A number of challenges are faced in terms of reimbursement of drugs, simultaneous co-infection with human immunodeficiency virus, access to allogeneic stem cell transplantation and, until recently, a lack of local recommendations appropriate for our setting. It is hoped that the newly published recommendations for the management of CML in South Africa will close many of the gaps in knowledge and practice and thus translate into better patient outcomes. Epidemiological data are limited and there is a need for more collaborative studies locally to elucidate issues such as incidence, prevalence and response to treatment. The challenges experienced in the management of CML and other cancers in the developing world are often economical and political in nature and require a comprehensive approach by clinicians, pathologists, health economists, medical insurers and policy makers if we are to find sustainable solutions.

Keywords:

• CML
• South Africa
• Treatment
• HIV

Introduction

The management of chronic myeloid leukaemia (CML) has undergone major changes over the last decade. Tyrosine kinase inhibitors (TKIs) have dramatically impacted patient outcomes with 70–80% of patients achieving a complete cytogenetic response or better. Nevertheless, about 25% of patients do not achieve complete cytogenetic response by 18 months and many patients lose their initial response over time due to resistance, non-compliance or intolerance and subsequent discontinuation of TKIs. Management of these patients is costly, requires treatment by experienced clinicians and appropriate laboratory facilities. Many gaps in knowledge and practice have been identified which may negatively impact patient outcome. Management of CML in South Africa is further complicated by a number of factors unique to our region. Some of these issues will be discussed here, as well as the actions taken to overcome them.

Epidemiology of CML

Studies from Europe and the United States of America have cited the incidence of CML at between 0·6 and 2 per 100 000/year. These data should be interpreted with caution as these are often based on early versions of the International Classification of Diseases for Oncology, which did not distinguish between BCR-ABL positive and negative CML, chronic myelomonocytic and ‘sub-acute’ myeloid leukaemia. These studies have also shown that CML increases with age with a male: female ratio of between 1·3 and 1·8.

South African incidence data for CML are lacking. A small study on 25 patients dealing with the demographic and clinical profile of CML in South African blacks was published in 1976.1 The last figures available from the South African National Cancer Registry dates from 2001 and do not distinguish between CML and other types of leukaemia. It showed a crude incidence of leukaemia in South Africa in 2001 of 2·08/100 000 for males and 1·44/100 000 for females, with 458 and 338 new cases reported in that year, for males and females, respectively. These data also showed a higher incidence of leukaemia in Asian and White patients, as compared to other ethnic groups, which may possibly reflect differences in socio-economic status and access to healthcare rather than true differences. Unfortunately, these data suffer from under-reporting as this is a passive, pathology-based registry. An effort is underway to update these data and notification of all newly diagnosed malignancies has recently become compulsory in South Africa, which should improve accuracy.

A recent, but unpublished study from the University of the Witwatersrand in Gauteng, South Africa, on 119 newly diagnosed patients with CML since 2002 showed a male: female ratio of 1·17, with a mean age of 42 years (range 16–88) (personal communication Professor Paul Ruff). Recently analyzed data from our group at the University of the Free State on 97 newly diagnosed patients from our region for the period 2000–2010, showed a male: female ratio of 1·04 and a median age of 38 years (range 2–84) (unpublished data). For patients older than 18, the median age was 44 years (range 18–84). These data are interesting, as it seems as if patients in South Africa present at a younger age than patients from more developed countries where the median age at presentation is about 50 years in clinical trials, and 10–20 years older in registry data. This probably reflects differences in the age structures of the populations rather than a real difference, as most of the developed countries show an aging of their populations, while this is not the case in most African countries. In South Africa, the life expectancy at birth in 2011 was estimated to be 54·9 years for males and 59·1 years for females. This may also explain the slightly lower male: female ratio seen in our population. Similar findings have been reported from other African countries., where median ages of 38, 35, and 39 have been reported for newly diagnosed CML patients in Nigeria, Kenya and Malawi, respectively.2^–4

Data on CML prevalence are limited and often combined with the general category of leukaemia. A French study has shown a progressively increasing annual prevalence rate (p.r.) since the introduction of TKIs with prevalence rates almost doubling from 1985 (p.r. of 5·8) to 2007 (p.r. of 10·4). Thanks to the effects of TKIs on survival, it has been estimated that there will be >250 000 patients living with CML in the USA by 2040.5

Epidemiologic information is important with regard to disease incidence, prevalence, diagnosis, prognostication and are key to healthcare planning and policy-making, especially dealing with conditions where available treatments are effective, yet expensive. Both incidence and prevalence data in South Africa are lacking and future studies and registry data collection should differentiate between CML, and more specifically, BCR-ABL positive CML and other leukaemias. The introduction of TKIs and the consequent improvements in survival, are likely to skew the prevalence data for the common leukaemia group, falsely increasing prevalence ratios. Similarly, the high costs of TKIs and the impact of their use on other health priorities, will have to be considered in an environment of increasing CML prevalence, especially taking into account the envisaged National Health Insurance plan for South Africa.

Access to Therapy

Access to TKI therapy has had a major impact on the successful management and increasing prevalence of patients living with CML. The cost of these treatments in the developing world has proved prohibitive for patients without medical insurance. In South Africa, TKIs remain unaffordable in the public healthcare setting which is accessed by about 80% of the population. The burden of human immunodeficiency virus (HIV), a high prevalence of other communicable and non-communicable diseases, as well as a focus on preventive and primary healthcare, has limited the availability of resources to access expensive cancer drugs. Out-of-pocket acquisition of these drugs remains out of reach for the vast majority of people in a country where the average total income per household is $789 per month. For Black South Africans who comprise the vast majority of the population, the average household income is only $399 per month. When the costs of the drug, monitoring, doctor’s fees and adverse event management are added, it becomes clear that this is only affordable to a select few patients with medical insurance. TKIs are available in some treatment centres through clinical trials, but patient literacy, language barriers, co-infection with HIV and living far from treatment centres have been challenges experienced related to patient inclusion in these.

These challenges are not unique to South Africa, but common to most of the developing world. Thankfully, patient assistance programmes are available that supply TKIs free of charge to poorer patients. The Glivec International Patient Assistance Program (GIPAP) has been set up as a partnership between Novartis and The Max Foundation. Since its inception in 2002, it has provided imatinib to more than 19 000 patients in 81 countries through a number of registered healthcare providers. At present about 500 patients in South Africa are active on the GIPAP programme. In total, approximately 2700 patients from sub-Saharan Africa were active on the GIPAP programme in June 2011 (personal communication Max Foundation). Similar, but smaller patient assistance programmes have recently been established in South Africa to facilitate access to nilotinib and dasatinib for patients with intolerance or resistance to imatinib. A concerning trend has been that patients with medical insurance have been refused therapy by insurers due to the cost of drug. As these patients do not qualify for GIPAP, these patients are falling through the gap and are not able to receive appropriate treatment. Better regulation and risk-sharing models are needed within the medical insurance industry to overcome this problem.

Recommendations for Treatment

CML management has undergone dramatic changes over the past decade. The success of TKIs has led to a paradigm shift in terms of the goals of therapy, the monitoring of disease and a simultaneous move away from allogeneic transplantation towards TKIs as the appropriate and acceptable first-line treatment. Treatment success relies on careful monitoring and timeous treatment adjustment to ensure that therapeutic goals are reached on time. Due to the complexities and cost of monitoring, expert management and specialized laboratory platforms are required. In South Africa, clinical expertise and laboratory facilities for cytogenetic and molecular monitoring are available, but limited to a few centres only and there is only one practising clinical haematologist per 2 million population.6 This means that patients often have to travel very long distances to be seen by a haematologist or oncologist. Unfortunately, they are then often treated by general physicians with little or no experience in the management of CML.

The European LeukemiaNet has published widely used recommendations for the management of CML.7 Because of the specific needs, circumstances and challenges of the South African healthcare environment, local experts from both the public and private sectors formed a CML expert panel that meets twice a year to discuss issues pertaining to CML in South Africa. This has proved to be a very useful platform for the sharing of ideas, research collaboration and education. The panel produced a set of recommendations for the management of CML in South Africa, which were published in November 2011.8 Imatinib is recommended as first-line treatment in newly diagnosed patients with CML, while dasatinib, nilotinib are deemed appropriate in cases of resistance or intolerance to imatinib.8 These recommendations also consider monitoring, stem cell transplantation, the management of HIV co-infection, intolerance to TKIs, pregnancy and breastfeeding.8 Thus, allogeneic stem cell transplantation for CML in South Africa is recommended in advanced phase CML, in patients achieving a second chronic phase, selected patients with imatinib failure, resistance due to a T315I mutation and in patients failing a second generation TKI.8 Data from the largest academic transplant programme have shown excellent results with allogeneic stem cell transplantation making use of ex vivo T-cell depletion followed by repletion with incremental dose donor lymphocyte infusions. Sixteen of 18 (89%) patients survived at a median of 855 days follow-up with very limited toxicity.9 Unfortunately, allogeneic stem cell transplantation is only offered in two out of the nine provinces which makes it difficult for the majority of patients to access this treatment.

Drug Intolerance

Adverse events to TKIs are generally manageable, but may lead to discontinuation of drug in a substantial portion of patients. Specific challenges noted in South Africa in terms of adverse events include the problem of depigmentation in dark-skinned individuals, the concurrent use of traditional medicines, herbs and foodstuffs that may affect drug metabolism and bioavailability and in some instances increase drug toxicity. In one study we found that a widely consumed relish, called chakalaka, may increase adverse events, which completely disappear when this foodstuff is discontinued.10

Co-infection with HIV

In 2011, the estimated overall HIV prevalence in South Africa was 10·6% with an estimated 5·38 million people living with HIV. In the age group between 15 and 49, 16·9% is HIV positive. Not only has the HIV/AIDS epidemic placed enormous pressure on the health care system, it has also led to many patients with malignancies, including CML, being co-infected with HIV. In contrast to certain other haematological malignancies, no pathopysiological association between the development of CML and HIV has been described and their simultaneous occurrence is thought to be co-incidental rather than causal. This disease combination poses unique challenges in terms of management and monitoring, such as high pill-burden in patients treated with both TKIs and cARV (combined antiretroviral therapy) which affects compliance; increased number of hospital visits which leads to an increased absenteeism from work; drug–drug interactions due to polypharmacy and shared metabolic pathways which may increase the number of adverse events; co-infection with tuberculosis, requiring treatment with CYP3A4 and CYP3A5 inducers, such as rifampicin, which may lead to subtherapeutic concentrations of TKIs (e.g. imatinib or nilotinib) metabolized by the same cytochrome. Efavirenz, a commonly used non-nucleoside reverse transcriptase inhibitor during first-line treatment of HIV, is also a strong inducer of CYP3A4 and may have similar effects on drug concentrations. These challenges may require increasing imatinib dosage by at least 50% in some patients, depending on response. Furthermore, HIV often leads to dysplastic changes in the bone marrow, which has led to concerns that this may negatively affect bone marrow reserve and the development of myelosuppression. In a recent series of 10 patients with CML co-infected with HIV from our unit, bone marrow dysplasia was present in 80% of patients, yet only one patient struggled with intractable neutropenia which precluded any increases in TKI dosage (unpublished data). A further concern is that the potential of some TKIs to cause hypogammaglobulinaemia and T-cell dysfunction may worsen the state of immunosuppression already present in HIV patients. We have not found this to be a clinical problem in practice, but longer follow-up is required. Despite the above, patients with CML and co-infection with HIV are often successfully treated with TKI therapy, but requires careful monitoring for adverse effects and response and motivation to comply.8

Disease Monitoring

Monitoring of disease response is key to treatment success in CML. Reaching goals within certain time frames and taking appropriate action early, when treatment fails or is suboptimal, is vital. Monitoring of CML by cytogenetic, real-time, quantitative polymerase chain reaction (RQ-PCR) and BCR-ABL mutation analysis is available in a number of centres that treat CML in South Africa. Where these are not locally available, samples are sent to reference laboratories for evaluation. If conventional karyotyping is not available or fails, fluorescent in situ hybridization (FISH) with dual colour, dual fusion translocation probes, is used. Once FISH positivity falls to below 10%, three-monthly RQ-PCR is used for further monitoring of disease response. Although FISH may detect BCR-ABL positivity where the Philadelphia chromosome is masked and in some variant translocations, it does not usually detect clonal evolution, is not validated to be used in this setting and may miss additional chromosomal abnormalities. Due to the high cost of BCR-ABL sequencing, our group has been studying the use of high-resolution melting curve analysis as a screening tool for the detection of BCR-ABL mutations.

Conclusion

Despite numerous challenges in a limited-resource setting, the management of CML in South Africa is very similar to that in developed countries, thanks to access to TKIs through patient assistance programmes, clinical trials and laboratory platforms that allow proper monitoring of disease. It is hoped that the newly published recommendations for the management of CML in South Africa will close many of the gaps in knowledge and practice and thus translate into better patient outcomes. There is a need for more collaborative studies locally to elucidate issues such as incidence, prevalence and response to treatment. It is prudent to remember that the challenges experienced in the management of CML and other cancers in the developing world are often economical and political in nature and require a comprehensive approach by clinicians, pathologists, health economists, medical insurers and policy makers if we are to find sustainable solutions.