Chemotherapy Medications in Sub-Saharan Africa: Availability, Pricing, Affordability, and Predictors of Quality.

INTRODUCTION
Cancer is a growing public health challenge in sub-Saharan Africa, with nearly 850,000 new cases and 560,000 deaths each year.1 Owing to population growth and aging, this is expected to rise to 1.5 million new cases and one million deaths by 2040.1,2 Access to effective and affordable treatment is essential to reduce deaths, prevent catastrophic health care costs, promote health equity, and contribute to the United Nations' Sustainable Development Goals to reduce premature mortality from noncommunicable diseases.3

CONTEXT

Key Objective

What is the availability, price, affordability, and characteristics that predict the quality of chemotherapy medications across the four countries: Cameroon, Ethiopia, Kenya, and Malawi?

Knowledge Generated

We observed stockouts of chemotherapy medications in public hospitals in three of four countries. Besides Malawi where medications are free when available in the public sector, chemotherapy medicines are unaffordable, costing many days of government worker salaries to treat cancer. On average, 2%-25% of medication cost ($4 US dollars [USD]-$214 USD) was spent on poor-quality medicines for treatment of head and neck, breast, or colorectal cancer in these countries. Price of the medication, public/private source, and medicine registration status were poor predictors of medicine quality.

Relevance

Patients seeking treatment for cancer can potentially spend their livelihoods on chemotherapy medications that may not be of legitimate quality. Better access to cheaper, quality-assured chemotherapy medicines is needed in sub-Saharan Africa.

Globally, nearly two billion people do not have access to essential medicines.4 This includes some cancer treatments which are increasingly in demand in sub-Saharan Africa because of the growing burden of noncommunicable diseases. Yet, sub-Saharan Africa faces inequities in cancer care and treatment compared with other regions. Restricted access to care, inadequate screening and diagnostic facilities, the absence of a well-structured and efficient cancer registry system, shortage of specialized medical staff, and lack of technical capacity are some of the reasons contributing to this inequity.5 Moreover, these disparities are further exacerbated by delayed diagnosis, having limited awareness of cancer, inadequate funding for cancer research, suboptimal vaccination uptake, and high costs for screening and treatment.5
Sub-Saharan Africa also faces high risks for having substandard and falsified medical products.6,7 The WHO defines substandard medical products as authorized medical products that do not meet their quality standards, specifications, or both,8 which arise because of poor manufacturing practices or inadequate quality control. Falsified medical products are those that deliberately misrepresent their identity, composition, or source.8 Weak regulatory systems, lack of access to affordable medicines, corruption, low consumer awareness and education, and supply chain complexity are some of the key drivers of substandard and falsified medical products.8,9 The WHO estimates that 1 in 10 essential medicines in low- and middle-income countries (LMICs) are substandard or falsified, while higher rates have been observed in Africa.6,7,10 There are limited data specifically assessing the incidence of chemotherapeutics being substandard or falsified. These medications could be a prime target as they tend to have elevated prices that could lead to higher profits for illicit manufacturers and distributors. The narrow therapeutic index of many chemotherapy medications increases the risk of health consequences to patients who unknowingly receive poor-quality medical products.
Understanding how available, affordable, and effective cancer medications are in sub-Saharan Africa is important to help identify and address gaps in chemotherapy access, improve treatment outcomes, and reduce cancer mortality. Evidence on the availability, price, and affordability of cancer medications on the continent is rising but remains limited.11-14 Data on the quality of chemotherapy medications in sub-Saharan Africa also remain limited because of the costs and challenges of following safety procedures required for testing. This study builds on findings from chemical testing of chemotherapy medications collected in four geographically and structurally diverse countries in sub-Saharan Africa—Cameroon, Ethiopia, Kenya, and Malawi. We examine the availability, price, and affordability of commonly used chemotherapy medications in these countries and assess characteristics that could predict substandard and falsified chemotherapy medications.

METHODS

Availability of Anticancer Medicines
Samples of seven medications—cisplatin, cyclophosphamide, doxorubicin, ifosfamide, leucovorin, methotrexate, and oxaliplatin—were collected in Cameroon, Ethiopia, Kenya, and Malawi between August 2023 and April 2024 by A.A., A.M.F., H.S.K., Y.S., and P.M.W. who have pharmacy or medical training and work at institutions providing oncology care in the study countries. Medication samples were directly collected from major national public hospitals that provided cancer care in three countries and collected covertly from public hospitals in one country. We recorded when any of the seven anticancer medications were available or not in stock at the time of collection. Medication samples were also collected from private community pharmacies that stock anticancer products where staff either presented medical prescriptions to obtain these medications in three countries or directly asked for specific medicines in one country. Medication sample data including the dosage, units, brand, manufacturer, country of manufacture, lot number, and price of the medicine were recorded in a digital database system (Artifacts VERIFY, Artifactsofresearch Inc, San Diego, CA). The country of manufacture was categorized as high-income countries (HICs) or LMICs based on World Bank country income classifications.15 The regulatory registration status of each sample was sought either from the country's regulatory authority or searched in lists of registered medicines. Sometimes the published lists were out of date where registration status could not be confirmed. At a minimum, registration meant a dossier was assessed by regulatory experts and the product was subsequently authorized to be marketed in the country. To contextualize findings on availability of anticancer medicines, we conducted a situational analysis where we described the procurement and supply chain of the medicines based on published literature and authors' experience providing oncological care in the study countries (Data Supplement).

Prices of Anticancer Medicines
Medication prices were recorded in the local currency and converted to US dollars (USD) using an exchange rate during the timeframe of sample collection.16 Out-of-pocket prices of the medication as well as the proportion of the cost that is subsidized by the government (in Ethiopia and Malawi) were recorded (see Data Supplement for details on subsidies). Where there were nonitemized receipts in Cameroon, we used prices for the same brand, dosage, and source in December 2024 as medication prices were stable over the year. Medication prices of paclitaxel and fluorouracil (FU), which were included in calculating the cost of treatment regimens but not sampled for quality tests in this study, were collected separately in October 2024 to January 2025 at hospital and community pharmacies in each country. We estimated the price per milligram of the active pharmaceutical ingredient (API) by dividing the stated milligram dose of the sample extracted from the package by the price of the medicine in USD.

Affordability of Anticancer Medicines
We estimated the total costs of anticancer medicines for the treatment of three common cancers from the perspective of the patient and government payer. The patient perspective reflects what a patient would have to pay for the medication at a public hospital or at a private pharmacy. The government payer perspective captures the portion of the medicine cost that the government pays for the patient when subsidizing medicines. We used the National Comprehensive Cancer Network's harmonized guidelines for sub-Saharan Africa for treatment of cancer by type.17 We examined the costs of medications for early-stage breast cancer, colorectal cancer (CRC), as well as head and neck cancer because treatments for these cancers involved many of the seven medications in our analysis.
For treatments of early-stage breast cancer and CRC which comprises multiple medicines, costs were estimated using a bootstrap method where we sampled from available country-specific prices of each medicine 1,000 times for the number of cycles needed and presented the average. Cost ranges are shown as the minimum and maximum total costs possible to pay for the entire regimen, demonstrating the range of costs if the patient were able to procure only the cheapest or the most expensive medicines at the time of their cancer treatment in each country. For head and neck cancer treatment, the average cost of cisplatin across three cycles was estimated.
The affordability of medicines to treat three cancers was assessed by calculating the number of days a person would have to work to pay for the total cost of the medication regimen. The salary of the lowest paid government worker from each country was used to estimate a daily wage using an assumption of 22 working days per month.18-21 The amount of days needed to pay for treatment is presented separately for a patient buying medicines from public and private sources where data were available.

Predictors of Substandard Anticancer Medicines
Collected medication samples were shipped to the University of Notre Dame in the United States for high-performance liquid chromatography (HPLC) testing. Details on sample storage, shipping, and HPLC materials and methods have been published previously.22 Results of the HPLC analysis reported a percentage of the API found in each sample. The API content for each sample was compared with the standard assay value limits set by the USP, and each sample was determined to have passed or failed HPLC on the basis of these limits. Testing results are reported for unique lots, where the percentage API content for tests conducted on samples with the same lot number were averaged across the lot.
Using these results and sample metadata, we ran a logistic regression to examine the factors associated with poor-quality anticancer medicines. Logistic regression is appropriate for modeling binary outcomes (pass/fail HPLC) and estimating the odds of failing HPLC based on other explanatory variables. The dependent variable in our model was the HPLC result (pass/fail) on the basis of API content. The primary explanatory variable was the price per milligram of each sample as a continuous variable. Other explanatory variables in the main model included whether the medicine was purchased from a public or private source, visual inspection results (pass/fail), the manufacturing location by country group (HICs or LMICs), and registration status of the medication (not registered/unknown or registered). We also ran the model with price per milligram as a categorical variable and examined other explanatory variables such as whether the medicine was purchased in a hospital or pharmacy, the study country, and specific medication. Separate logistic regressions were run for individual APIs to examine the association between medicine prices and HPLC result, controlling for other explanatory variables.
Analyses were conducted using Python 3.7.13 (stats models package). Continuous variables were standardized, and categorical variables were dummy-coded using one-hot encoding, with the reference category omitted. To account for potential collinearity, variance inflation factors (VIFs) were examined and variables with excessive multicollinearity (VIF > 10) were considered for removal.23 We report the marginal effects as an expected change in the probability of HPLC failure given a one-unit change in each of the predictor variables. Statistical significance was assessed at P values < .05.
On the basis of the observed probability of obtaining a medicine that failed quality testing, we also estimated how much money could have been spent on poor-quality medicines for each of the three cancer regimens. We estimated a 95% CI around the failure rate of cyclophosphamide, doxorubicin, cisplatin, leucovorin, and oxaliplatin using the Wilson score interval method24 and multiplied this by the average cost of the medicine across all treatment cycles across countries.

RESULTS

Availability of Anticancer Medicines
We collected a total of 329 medicines across four countries, including samples of cisplatin (n = 51), cyclophosphamide (n = 42), doxorubicin (n = 50), ifosfamide (n = 25), leucovorin (n = 37), methotrexate (n = 78), and oxaliplatin (n = 46). Most (81%) samples were collected from private pharmacies. In Kenya, Cyclophosphamide, doxorubicin, and cisplatin were only purchased from private facilities. Some anticancer medicines were not available at public hospitals during our data collection period. Leucovorin injectables were not available because of stockouts in public hospitals in Cameroon. In Malawi, cisplatin was stocked out in public hospitals. In Cameroon, methotrexate was not available in public hospitals during the study period.

Prices of Anticancer Medicines
We found that the purchase price of the seven anticancer medications in this study ranged from $0.03 USD for a locally manufactured leucovorin tablet in Ethiopia to $61 USD for ifosfamide. The most expensive cost paid per vial of medicine came from private sources for ifosfamide 2,000 mg packed with two vials of Mesna in Ethiopia which cost $61 USD, and for ifosfamide 1,000 mg purchased for $58 USD in Cameroon. In Kenya and Malawi, the highest prices paid in this study were for vials of oxaliplatin 100 mg which cost $52 USD in both countries, where they were obtained from a public hospital in Malawi and from a private pharmacy in Kenya. Table 1 shows the average prices for common dosage forms of the medicines across all countries. The average prices per unit were similar across sites where oxaliplatin ($28.53 USD-$48.99 USD for 100 mg) and ifosfamide ($15.13 USD-$48.99 USD for 1,000 mg) were found to have the highest average cost per unit.

Affordability of Anticancer Medicines
The costs of medicines for treating three common cancers are presented in Table 2. In Malawi, where patients are fully subsidized for cancer treatment in public hospitals, the cost to the patient was $0 USD for breast and CRC treatments. However, because of medicine stockouts, the only option for treating head and neck cancer during the study period was to obtain privately sourced medicines out of pocket, costing an average of $254 USD (min-max, $138 USD-$277 USD). In Ethiopia, the price to the patient of medicines for a four-cycle breast cancer treatment cost on average $79 USD ($65 USD-$95 USD) and three cycles of cisplatin for head and neck cancer cost $66 USD ($18 USD-$162 USD). Those prices represent 50% of the entire costs of medicines, where the other half was paid for by the government through a subsidy.
Because of stockouts noted above, the price for some public regimens represents a combination of medicines available at public facilities, supplemented with medicines from private suppliers. This is shown in the cost for CRC treatment in Ethiopia where leucovorin injectable was often unavailable in public hospitals, forcing the patient to obtain the medication from a private source, for which they must pay the full price. Cost to the patient were substantially higher for medicines to treat breast cancer in Kenya ($982 USD, $489 USD-$1,758 USD) and Cameroon ($241 USD, $137 USD-$408 USD), where no subsidies were available. However, the total average cost of each treatment regimen did not differ widely by country. Average costs of treatment regimens were consistently more expensive in private community pharmacies compared with prices at major national public hospitals. The maximum end of the ranges demonstrates the prices when all medicines were purchased from the most expensive sources. For example, it cost nearly $2,000 USD to treat CRC in Kenya or Ethiopia when purchasing from private sources.
Figure 1 demonstrates how many days of pay at the lowest government salary a patient would have to incur to afford the medications for cancer treatment. The infusional fluorouracil, leucovorin, and oxaliplatin regimen for treating CRC was the least affordable, where patients in Ethiopia would have to contribute 869 days (nearly 2.5 years) of their salary to purchase those medicines from a private source or 757 days (nearly 2.1 years) if they received treatment in public hospitals where they could receive a 50% subsidy only on oxaliplatin and FU when in stock. CRC treatment can be affordable in Malawi if a patient received the medicines from a public hospital, which would cost them 0 days' wages. However, if the medicines were stocked out, they would have to work 233 days (7.7 months) to pay for medicines purchased from a private source. Comparatively, Cameroon had the most affordable medication cost for patients for head and neck cancer, where it cost nearly a month's salary for the lowest paid government worker to pay for three cycles of cisplatin. However, if a patient were forced to source their medicines from private pharmacies, cisplatin costs over 2.5 months of salary in Cameroon.

Predictors of Substandard Anticancer Medicines
Quality testing with HPLC was conducted on 251 of the samples from 191 unique lots. Of the 191 lots, 32 (16.7%) did not meet USP standards for API content and 25 (13%) were borderline assay values, within 2% of the API cutoffs. To be conservative, we considered borderline results as passing HPLC tests in the regression analysis. We found the following numbers of failed samples by API: cisplatin two fails (6.7% of n = 30), cyclophosphamide 11 fails (35.5% of n = 31), doxorubicin nine fails (25.7% of n = 35), ifosfamide zero fails (0% of n = 4), leucovorin one fail (5.6% of n = 18), methotrexate nine fails (19.1% of n = 47), and oxaliplatin zero fails (0% of n = 26). Visual inspection of the medicine and its packaging resulted in 14 failures, six of which also failed HPLC testing. One sample of doxorubicin from Ethiopia had incomplete data and was excluded from the regression analysis.
Table 3 presents the marginal effects of the logistic regression on the 191 lots. The study country, specific medication, and whether a medicine was purchased in a hospital or pharmacy were removed from the model because of multicollinearity. We found that the medicines being manufactured in a LMIC was associated with a 36% increase in failing HPLC compared with being manufactured in a HIC, holding all else constant (P = .013). Per milligram price of the medicine, public or private source of medicines and regulatory registration status were not statistically significantly associated with the probability of failing HPLC. The medicine failing visual inspection were also not statistically significantly was associated with a 25% increase in the risk of failing HPLC (P = .038).
Table 4 presents the logistic regression results with marginal effects for three APIs with large enough sample size for analysis: cyclophosphamide, doxorubicin, and methotrexate. Variables that might have affected prices such as the study country, manufacturing country, and regulatory registration status were tested in this model but removed because of multicollinearity and small sample size. We found that prices were not statistically significantly associated with HPLC failure after controlling for the source of the medicine. Only for doxorubicin, private procurement was predictive of HPLC failure after controlling for price.
For each medication, we estimated 95% CIs around the probabilities of obtaining medicines that failed HPLC. On the basis of the average chemotherapy costs across countries, we estimate that on average across the population $86 USD (95% CI, $50 USD to $134 USD, 9%-25% of medicine costs) was spent on substandard and falsified medicines for treatment of early-stage breast cancer, $32 USD (95% CI, $6 USD to $214 USD, 0.5%-17%) for CRC, and $25 USD (95% CI, $4 USD to $49 USD, 2%-21%) for head and neck cancer (Fig 2).

DISCUSSION
We found that medicines for cancer treatment are unaffordable in sub-Saharan Africa, especially when treatment requires a combination of costly APIs. Only in Malawi, where treatment is fully subsidized, could a patient working for minimum wage access cancer medicines without an excessive burden if they are available at public facilities. This finding is consistent with previous studies that found that anticancer medicines were less affordable in LMICs compared with HICs using both patient income and gross domestic product approaches, despite the unit prices of medicines being significantly higher in HIC settings.13,25 Despite the large cost burden of anticancer medicines on patients and their families, we found that the quality of chemotherapy medicines is not guaranteed, which threatens medication effectiveness and patient survival. These results highlight the importance of interventions to improve medicine access such as supply chain management to ensure medicine availability at public hospitals, government subsidies to make chemotherapy medications affordable, and chemical testing through postmarket surveillance to safeguard the quality of chemotherapy medications.
Our results illustrate how closely intertwined availability, affordability, and quality of chemotherapy medicines are. Stockouts at public hospitals threaten not only availability but also the affordability and quality of medicines. Even in countries such as Ethiopia and Malawi where subsidies have aided the affordability of anticancer medicines, they become unaffordable when there are stockouts in public facilities. Lack of availability pushes patients to seek medicines from the gray market, where patients may face higher risks of obtaining substandard or falsified medicines. The causes of stockouts are multifaceted; however, within the context of the study countries, procurement and supply management issues were key drivers of stockouts, especially within public sector facilities.14,26-29 Central medical stores need better tendering processes that could guarantee medication supply and assure quality. Moreover, better supply chain management predicting the demand for chemotherapy medications could minimize shortages and wastage. Subsidizing the cost of treatment at public hospitals would improve affordability and may encourage earlier care seeking that would improve patient survival. Availability, affordability, and quality are all essential dimensions of medicine access, in addition to accessibility and acceptability.30 Future studies should examine the impact of various policies on these dimensions to improve overall access to chemotherapy medicines in sub-Saharan Africa.
Price of a chemotherapy medication was found to be a poor indicator of quality. Moreover, we found that public/private source and registration status of medications were also not good predictors of medicine quality. This has implications for policy because it implies that substandard and falsified medicines could be cheap or expensive, prevalent in both public and private sectors, and that country requirements to register medicines are not helping to counter the presence of substandard and falsified medicines in the market. We found that medicines manufactured in LMICs were significantly associated with increasing the probability of failing an HPLC test. However, a large portion of anticancer medicines available in the study countries (and 80% of those included in this sample) are manufactured in India. The country of manufacturer may be a risk, but it is not sufficient to rule on quality. To address source country risks, collaboration is essential between governments and global partners to ensure a harmonized approach to regulation and enforcement action against exporters who do not meet quality standards. Similarly, we found that visual inspection is only good at predicting some quality issues, where six of the 14 visual inspection failures were found to have insufficient API. Without chemical testing of the medicines, these medicine characteristics were not sufficient as predictors to determine whether a patient should receive a medicine or not.
Currently, the only reliable way to test if an anticancer medicine is of sufficient quality is to conduct HPLC testing, which is expensive, time intensive, and consumes the entire product where the medicine cannot then be given to the patient after the test. Monitoring the quality of anticancer medicines in LMIC settings is complicated by the need to have personal protective equipment, trained manpower, and other demands to perform quality of anticancer medicines besides the additional set up of sophisticated laboratory equipment. However, the emergence of alternative field screening technologies, such as the chemotherapy paper analytical device (chemoPAD), offers simpler, efficient, and cost-effective ways of screening anticancer medicines for quality at the point of care.31-34 Two studies conducted in Ethiopia by our team using the chemoPAD documented the presence of substandard and falsified cisplatin and doxorubicin products in tertiary public university teaching hospitals and showed that the chemoPAD had excellent sensitivity and specificity.31,32 Although the chemoPAD cannot replace HPLC and pharmacopeial methods, its ease of use and low cost can allow greater expansion of sampling and testing of anticancer medicines, as part of a risk-based approach to postmarketing surveillance.35,36 Clinicians and pharmacists could use the chemoPAD and other screening technologies to assess the quality of anticancer medicines, beyond just visual inspection, while national medicine regulatory authorities should focus limited resources on more targeted confirmatory testing using HPLC and overall regulatory strengthening.36,37
Our results call for an urgent need to chemically test chemotherapy medications in sub-Saharan African countries to ensure that patients are receiving quality medications and for greater investments in regulatory strengthening and quality assurance for chemotherapy medications. This would likely require additional training, lab equipment, and financial resources to strengthen national medicine regulatory authorities to conduct testing and removal of substandard and falsified chemotherapy medications to improve patient safety and survival. Medication availability would also have to be ensured when conducting medicine quality testing so that medicine access would not be inversely affected by ensuring quality.
The WHO reports that the availability of medicines for cancer is 32.0% in LMIC and 57.7% in low-income countries.38 Other studies have reported the scarcity of anticancer medicines in sub-Saharan Africa.25,39 A few studies that document the availability of anticancer medicines had reported that availability was 42% for Rwanda,40 50% in Tanzania,41 34.8% for low-priced generic anticancer medicines and 2.8% for the originator branded anticancer medicines in Ethiopia.26 Anticancer medicine availability was 13.1% for generics and 5.4% for original brands in the public sector and 10.6% for generics and 2.5% for originator brands in Ghana.42 Improving the availability of chemotherapy medications in sub-Saharan Africa is essential to reduce cancer mortality.
Lack of access to affordable and quality anticancer medicines in sub-Saharan Africa is a public health threat. The region has a disproportionate and growing burden of cancer mortality. The annual number of cancer cases is projected to increase from 727,000 to 1.4 million between 2020 and 2040, whereas annual mortality will more than double from 484,000 to 967,000 during this same time period.43,44 Moreover, governments are spending their limited resources sometimes on medicines that do not meet quality standards. Greater attention should be placed on improving access to quality-assured chemotherapy medications in sub-Saharan Africa to counter the growing burden of cancer in the region and reduce premature mortality. Organizations such as the Africa CDC and Prevention, the African Medicines Agency, and the African Union Development Agency—New Partnership for Africa's Development can play a key role in elevating the burden of substandard and falsified chemotherapy medicines as a policy concern and in providing technical assistance to national medicine regulatory authorities and Ministries of Health, similar to Africa CDC's initiative to improve regulation and access to quality assured diagnostics.45
Our analysis involves a number of key limitations. The analyzed sample of medicines does not represent a random sample, and the HPLC failure rate of these medicines should not be used to assert the overall prevalence of substandard and falsified anticancer medicines. The results of the analysis come from four countries in sub-Saharan Africa and are not generalizable to the region as a whole. The sampling methods were a mixture of covert and overt sampling across the study countries, which has affected what kind of medicines were accessed from public and private sources, resulting in large numbers of medicines coming from private sources where prices are higher. We therefore present our costs separately for public and private facilities to document cost differentials. Because our focus was on medications, we present only the medication costs and do not include the costs of hospitalization or other care-seeking costs. Efforts were made to sample a large variety of medicines, doses, and brands, but the sample analyzed here is still relatively small which could skew the results of the price and regression analyses. However, as prices and manufacturers seen across the four countries were fairly consistent, we believe the sample presents a reasonable snapshot of anticancer medicines that a cancer patient in our study countries would have had access to within our study period. To assess medicine affordability, we used the wages of the lowest paid government worker in each country which may not be representative of what an average person can afford. However, we believe this represents a conservative estimate of affordability in countries with large informal sectors and without an enforced minimum wage. Finally, our estimates of spending on substandard and falsified medicines do not represent population-level costs. Future analyses should examine the economic impact of substandard and falsified chemotherapy medications across cancer patients in these countries.
In conclusion, despite these limitations, our findings are important in describing the availability, pricing, affordability, and predictors of quality of anticancer medicines in sub-Saharan Africa. Without the help of government subsidies, patients seeking chemotherapy medicines will potentially spend their livelihoods to pay for treatment. When these medicines are substandard or falsified, the health and economic burden on the patient and their family can be devastating. Access to chemotherapy medicines in sub-Saharan Africa needs to be strengthened, and safe and inexpensive ways to test the quality of chemotherapy medications are needed to reduce premature mortality from cancer.