Chemotherapy-Induced Toxicities in Pediatric Cancer Patients: A Cross-Sectional Survey Using a Child- and Caregiver-Reported Outcome Tool in Tanzania

Introduction

Chemotherapy remains the cornerstone of treatment for most pediatric cancers, acting by inhibiting cell division or directly killing malignant cells. Globally, over 400,000 children and adolescents aged 0–19 years are diagnosed with cancer annually.¹ In Tanzania, an estimated 4000 to 4500 children are diagnosed each year.² Common childhood cancers in the country include leukemia, brain tumors, lymphomas, and solid tumors such as neuroblastoma and Wilms’ tumor. Many of these cancers are potentially curable with appropriate combinations of chemotherapy, surgery, and radiotherapy.³ However, survival rates for pediatric cancers in Tanzania and other low- and middle-income countries remain substantially lower than those in high-income settings, largely due to late presentation, limited diagnostic capacity, and resource constraints affecting access to chemotherapy and supportive care medicines for prevention and management of chemotherapy-induced toxicities (CITs).⁴

Although chemotherapeutic agents significantly improve cure and survival rates, they are also associated with moderate to severe side effects known as adverse drug reactions (ADRs), which can reduce children’s quality of life.⁵ An ADR refers to any unintended and harmful effect occurring at doses used for prevention, diagnosis, or treatment.⁶ In the context of chemotherapy, these ADRs are referred to as CITs, which result from the drugs’ harmful effects on healthy, rapidly dividing cells.⁷ Studies have shown that, regardless of individual risk factors, two out of three pediatric cancer survivors will develop at least one chronic health condition related to treatment.⁸ Nearly 60% of childhood cancer survivors face potentially life-threatening complications later in life.⁸ Early recognition and management of these toxicities during treatment are therefore essential to prevent or lessen such long-term complications and to improve survival and quality of life. Common CITs include gastrointestinal symptoms (eg, nausea, vomiting, diarrhea, and anorexia), dermatologic issues (eg, rash, urticaria), systemic reactions, and central nervous system effects.⁹ Additionally, up to 40% of children undergoing chemotherapy may experience severe CITs.¹⁰

While CITs are a specific subset of ADRs, their frequency and severity are often much higher than ADRs observed with routine pharmacotherapy because chemotherapeutic agents are inherently cytotoxic and affect both malignant and healthy tissues.¹¹ Previous studies have reported that the incidence of ADRs among pediatric inpatients ranges from 0.4% to 10.3%,¹² whereas CITs may affect more than 80% of children receiving chemotherapy.¹³ In studies from Ethiopia and India, more than half of pediatric oncology patients experienced at least one CIT, with gastrointestinal and hematologic toxicities being the most prevalent.^14,15 These findings show the need for surveillance, monitoring and managing CITs in low- and middle-income countries, where reporting systems are limited.

Management of CITs depends on the specific symptoms reported. Fatigue is commonly managed through adequate rest, nutritional support, and treatment of contributing factors such as anemia or dehydration. Antiemetic medications are used to relieve nausea and vomiting, while medical evaluation is recommended for severe symptoms such as high fever, bleeding, mouth sores, diarrhea, constipation, or changes in skin color. Key strategies for CITs management include strict adherence to prescribed medications, ensuring adequate hydration, and promptly reporting complications to healthcare providers.¹⁶ Effective management requires both objective clinical assessments and subjective patient input. While clinicians rely on observation and diagnostic tests to detect CITs, child and caregiver reports provide vital insights into the lived experience of treatment-related side effects.¹⁷ Therefore, this study aimed to assess the types, severity, and determinants of child- and caregiver-reported CITs among pediatric cancer patients at Bugando Medical Center (BMC) using a culturally adapted, Swahili-translated version of the Pediatric Patient-Reported Outcomes version of the Common Terminology Criteria for Adverse Events (Ped-PRO-CTCAE). This study represents the first documented use of the Swahili-translated Ped-PRO-CTCAE tool in Tanzania and one of the earliest applications of this instrument in Sub-Saharan Africa.

Methodology

Study Area

This study was conducted at BMC, a consultant, university teaching, and referral hospital located in Mwanza, northwestern Tanzania. BMC serves as the Lake Zone’s main referral facility and is the second-largest cancer center in the country. The cancer center at BMC, has a total bed capacity of over 110, with approximately 20 to 30 beds designated for pediatric oncology patients. On average, 15–25 pediatric oncology patients are admitted each month for diagnosis, chemotherapy, or supportive care. The center diagnoses more than 200 new childhood cancer cases annually and provides services to the Lake Zone region, which has a catchment population of over 16 million people.¹⁸ The site was selected for this study because of its large patient volume, availability of specialized oncology services, and established experience in pediatric cancer management.

Study Design and Project Duration

A descriptive cross-sectional study design was employed to assess the types, severity, and determinants of CITs among pediatric cancer patients. This design was appropriate because it allowed the collection of information on toxicities and related factors at a single point in time among children actively receiving chemotherapy, providing a clear snapshot of the prevalence and patterns of CITs within the study population. The study was conducted over six months, from February to July 2023.

Study Population

The study population consisted of all pediatric cancer patients who had received at least one cycle of chemotherapy at BMC. For children unable to self-report (typically those under seven years of age), data were obtained from their primary caregivers (parents or guardians).

Study Criteria

Inclusion Criteria

Pediatric cancer patients aged from birth to 18 years who were in the active phase of chemotherapy treatment and had received a chemotherapy cycle within the preceding seven days.

Exclusion Criteria

Pediatric patients receiving only palliative care or with unconfirmed cancer diagnoses at the time of recruitment were excluded. In addition, children or caregivers with conditions that could compromise accurate symptom reporting, such as severe cognitive impairment, altered consciousness, or critical illness, were not eligible for participation.

Sample Size and Sampling Procedure

The institutional pediatric oncology database has records of over 800 patients since its establishment. At the time of the study, approximately 170 pediatric patients were actively receiving treatment. Using Yamane’s formula for sample size estimation at a 95% confidence level and a 5% margin of error:

n = N / [1 + N(e)²]

where n is the sample size, N is the population size (170), and e is the margin of error (0.05). The required sample size was calculated to be 119, and a total of 120 participants were successfully recruited. A convenience sampling technique was employed due to logistical constraints and the difficulty of identifying all eligible patients within the study timeframe. While this approach may limit generalizability, it was practical for the study setting. To enhance representativeness and minimize selection bias, recruitment was conducted across different treatment settings (inpatient, outpatient, and daycare units) and included participants of varying ages, sexes, and cancer types as they presented for care during the study period.

Data Collection Tools

Several methods are available for identifying CITs, including medical record reviews, laboratory data, medication charts, electronic ADR reporting systems, clinical rounds, and interviews with patients, caregivers, or healthcare providers. While some of these approaches generate objective data, not all are validated or standardized for pediatric use.¹⁹ In this study, we used the National Cancer Institute’s PRO-CTCAE, a validated tool designed to capture symptomatic toxicities as directly reported by patients. Its pediatric version, the Ped-PRO-CTCAE, is intended for use by children aged seven years or older, while a caregiver version is available for younger or non-verbal children. The tool assesses the frequency, severity, and impact of toxicities on daily life.²⁰

The Swahili version of the Ped-PRO-CTCAE used in this study was translated and culturally adapted at BMC following the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) guidelines to ensure conceptual equivalence with the original English tool. The process involved independent forward translation by bilingual experts, reconciliation, blind back-translation, and harmonization by a multidisciplinary panel. Linguistic validation was achieved through three rounds of cognitive interviews with pediatric patients and caregivers until ≥ 80% comprehension was reached. The adaptation process also included review by pediatric oncologists, linguists, and cultural experts to ensure the translated items reflected Tanzanian children and caregivers’ local cultural context and health literacy.²¹ The final version demonstrated high clarity and cultural appropriateness and has been validated for use in Swahili-speaking pediatric oncology populations.²¹

Data Collection Procedures

Two pharmacists who had worked in the oncology department for at least six months were recruited as data collectors. Before data collection began, they received intensive training and orientation on the study objectives, the contents of the questionnaire, and ethical principles of interviewing pediatric participants and their caregivers. The training included a pilot exercise conducted with a sample of 15 participants who were not part of the final study to ensure consistency and minimize interviewer bias. Data were collected at BMC’s pediatric oncology unit from both inpatients and outpatients, including children receiving chemotherapy during daycare or ward-based treatment. Interviews were conducted in a quiet area of the oncology unit either while patients were waiting to see the oncologist or shortly after their chemotherapy sessions, depending on their clinical condition and availability. Each interview lasted approximately 15–25 minutes and was administered face-to-face in Swahili by one of the trained data collectors. The Ped-PRO-CTCAE uses a 7-day recall window, enabling participants or caregivers to report toxicities experienced within the preceding week of chemotherapy. Children aged seven years and above completed the Ped-PRO-CTCAE questionnaire themselves, rating the frequency, severity, and impact of CITs on daily activities. For younger children or those who can not express themselves, including infants and toddlers, caregivers completed the caregiver-proxy version of the tool. Caregivers were guided by trained interviewers to interpret and grade each symptom based on observable indicators such as changes in appetite, crying, sleeping patterns, play activity, and general behavior. For participants below two years of age, caregiver responses focused on physical and behavioral cues rather than verbal expression. Interviewers ensured consistency in grading by clarifying examples during interviews and verifying reported symptoms with clinical notes and medication charts where possible.

Data Analysis Procedures and Statistical Analysis

Collected data were first entered into Microsoft Excel 2013 for initial cleaning, coding, and organization, and then imported into STATA (TX: StataCorp LLC) version 15 for analysis. Categorical variables are presented as frequencies and percentages using tables and a figure, while continuous variables are reported as means with standard deviations or as medians with interquartile ranges, depending on their distribution. The severity of CITs was assessed using the National Cancer Institute’s Common Terminology Criteria for Adverse Events (NCI CTCAE) version 5.0. Each toxicity was graded from 1 (mild) to 5 (death related to toxicity), based on clinical descriptions of severity. Grading was performed by two trained pharmacists who collected the data, and all entries were reviewed and verified by the principal investigator to ensure consistency and accuracy. Associations between CITs and independent variables (covariates including age, sex, cancer type, number of chemotherapy drugs, and vincristine-based regimen use) were initially examined using contingency tables with Chi-square or Fisher’s exact tests, as appropriate. Multivariate logistic regression was performed to account for potential confounding factors or variables. Odds ratios (ORs) with 95% confidence intervals (CIs) were calculated, and a p-value < 0.05 was considered statistically significant. Moreover, to avoid misinterpretation of severity data, CITs were analyzed at two complementary levels. First, symptom-level severity counts were summarized across all organ systems (eg, total grade 1, grade 2, and grade 3 events). Second, for regression analysis, a patient-level binary variable (ToxBinary) was constructed, classifying a participant as having ≥ Grade 2 toxicity if at least one reported symptom reached grade 2 or higher. Because participants reported multiple symptoms with varying severities, the total number of symptom-level events does not equal the number of patients classified as having moderate or severe toxicity. For patient-level severity reporting, the highest CTCAE grade across all symptoms reported by each participant was used to determine the distribution of severity grades among the 120 participants.

Results

Sociodemographic Characteristics of Participants

A total of 120 participants receiving chemotherapy were enrolled in the study. The majority were male (n = 67, 55.8%). The participants’ ages ranged from birth to 18 years, with a median of 6 years (interquartile range [IQR]: 1–18 years) and the largest proportion in the (3–6) years age group (n = 47, 39.2%). Body weight ranged from 2 kg to 70 kg, with a median of 19.0 kg (IQR: 9–55.7 kg) (Table 1). Overall, the cohort was predominantly male and skewed toward younger children, with most participants below 7 years of age.

Table 1 Sociodemographic Characteristics of Participants (N = 120)

Clinical Characteristics of Participants

Twenty-one different cancer types were recorded, with the most frequently occurring being nephroblastoma, acute lymphoblastic leukemia, Hodgkin lymphoma, and retinoblastoma (Figure 1). For participants with solid tumors (78.3%, n = 94), staging was applied using disease-specific pediatric oncology systems. Most were stage II (42.6%, n = 40). Participants with hematologic malignancies (16.7%, n = 20) were classified by risk, while brain tumors (5.0%, n = 6) were classified by grade. Most brain tumors were low grade (83.3%, n = 5) (Table 2).

Table 2 Staging, Risk, and Tumor Grade Among Participants

Figure 1 Types and frequencies of cancer among participants (n = 120).

Furthermore, twenty-one anticancer agents were used, with (68%, n= 82) prescribed in combination regimens. The most frequent combinations were vincristine, carboplatin, and etoposide (CEV) (10.8%, n=13) and oral etoposide with cyclophosphamide (ECO) (10.0%, n=12) (Table 3). Chemotherapy regimens were also categorized by the number of agents used. Most patients received two drugs (45.8%, n=55), followed by three (35.0%, n=42), one (12.5%, n=15), and four (6.7%, n=8). Vincristine-based regimens were used in (52.5%, n = 63) of patients.

Table 3 Chemotherapy Regimens Prescribed to Participants (N = 120)

Supportive Care Medications and Management of Chemotherapy-Induced Toxicities

Medical interventions were provided to alleviate CITs based on the type and severity of symptoms. Supportive care included the use of antiemetics, corticosteroids, gastroprotective agents, and other adjunctive therapies. Among the 120 participants, 32.5% (n = 39) did not receive any supportive care medication. Of those who did, 34.2% (n = 41) were given ondansetron alone, 27.5% (n = 33) received ondansetron plus another agent (eg, prednisolone, dexamethasone, diphenhydramine, omeprazole, pantoprazole, or famotidine), and 5.8% (n = 7) were prescribed other supportive drugs such as allopurinol, filgrastim, or ursodiol.

Chemotherapy-Induced Toxicities Among Participants

All participants reported experiencing at least one chemotherapy-induced toxicity (CIT), with severity ranging from mild to severe. Across all organ systems, a total of 692 individual CIT events were reported by the 120 participants. Because participants could report multiple symptoms with different severities, event counts exceeded the number of participants. The leading reported toxicities were alopecia (83%), taste changes (74.2%), vomiting (44.2%), and fatigue (30%). Less frequent but clinically relevant toxicities included cough (23%), mouth sores (19.2%), headaches (16%), and dry skin (13.3%). Table 4 summarizes the frequencies and percentages of all reported CITs, categorized by organ/system.

Table 4 Patient-Reported Chemotherapy-Induced Toxicities Among Participants (N = 120)

The Severity of Chemotherapy-Induced Toxicities Among Participants

For severity summarization at the patient level, each participant was assigned a highest CTCAE grade. Based on this classification, 54% (n = 65) of participants had a maximum grade of 1, 43% (n = 52) had a maximum grade of 2, and 3% (n = 3) had a maximum grade of 3. Participants also reported varying degrees of impact on daily activities: 48% (n = 58) indicated that some activities were affected, 40%, (n = 48) reported that many activities were affected, and 12%, (n = 14) reported that nearly all activities were affected.

Associations Between Chemotherapy-Induced Toxicities and Independent Variables

To account for potential confounding factors or variables, binary logistic regression analyses was conducted to identify factors associated with experiencing ≥ Grade 2 CITs in any organ system. Independent variables included sex, age group, cancer classification (solid vs hematologic), number of chemotherapy drugs, and whether the regimen contained vincristine. In unadjusted analyses, both solid and hematologic malignancies showed significantly higher odds of experiencing ≥ Grade 2 CITs (p = 0.031 for both). After adjustment for all covariates, cancer classification remained a significant predictor of toxicity: solid tumors were associated with a 7-fold higher odds of ≥ Grade 2 CITs compared with hematologic cancers (AOR = 7.42, 95% CI 1.02–53.79, p = 0.047). There were no statistically significant associations between toxicity and sex, age group, number of chemotherapy drugs, or vincristine-based regimens. The adjusted odds ratio for sex (AOR = 0.92, p = 0.893) was close to 1, indicating no meaningful difference in toxicity risk between males and females. The complete results of both unadjusted and multivariable logistic regression analyses are summarized in Table 5.

Table 5 Significant Associations Between Chemotherapy-Induced Toxicities (CITs) and Independent Variables Among Participants (N = 120)

Discussion

This study found that CITs were common among pediatric cancer patients, with most participants experiencing multiple mild to moderate toxicities, primarily gastrointestinal and cutaneous in nature. These findings highlight the ongoing challenges of managing chemotherapy-related side effects in children within resource-limited settings. In many cancer centers across Africa, pediatric patients are often treated alongside adults, making individualized follow-up and systematic monitoring of treatment responses difficult.²²

In the current study, a total of 120 pediatric cancer patients receiving chemotherapy were recruited over a period of six months. This sample size was similar to the study done at another Pediatric Oncology Unit (POU), in Muhimbili National Hospital (MNH), Dar es Salaam, Tanzania, whereby a total of 123 pediatric cancer patients were eligible for analysis.²³ Both BMC and MNH are the current biggest pediatric cancer centers in the country. While MNH serves the eastern, central, and some southern parts of the country, BMC serves the western, northern, and some central parts of the country. This may suggest that these centers share an equal burden of the increasing prevalence of pediatric cancers in Tanzania.

Most participants were in the age group of 3–6 years (39.2%), which could be a reason why most of them had nephroblastoma and Acute Lymphoblastic Leukemia (ALL), which have a peak incidence age of 2 to 4 and 2 to 6 years, respectively. This is similarly reported by another study done at the Kilimanjaro Christian Medical Center (KCMC), a third pediatric cancer center in Tanzania, but different from the results observed in another study done at Jimma University, Southwest Ethiopia, where childhood cancer was mostly diagnosed in children between 6 and 11 years of age.^4,14 Also, in the present study, most participants were male, consistent with findings from Ethiopia, but contrary to a study conducted in India.^24,25 This may be due to the difference in pediatric cancer patterns between East Africa and India. Similarly, the cancer diagnosis rates in the current study were similar to those in a study in northern Tanzania and Ethiopia.^4,26 Nonetheless, the general demographics of the participants are consistent with other studies in the Sub-Saharan Africa region.^27,28 The difference shown by the studies from other regions may be due to variations in genetics and environmental exposure between geographic environments.

One of the determinants of the type and selection of cancer treatment is the severity of the disease, measured by the stage, risk, or grade of the disease. Many epidemiological studies in low and middle-income countries (LMICs) on pediatric cancers reported the high proportions of participants presenting at the hospital with advanced disease, and this has been a reason for poor outcomes.²⁹ Most participants in the current study had stage II cancer (33.3%, n=40), followed by stage IV cancer (23%, n=27), contrary to a retrospective study in a tertiary hospital in northern Tanzania, which reported that over 75% of participants had stage III or above.⁴ The difference could be due to differences in methodologies between the two studies. While the current studies collected data from participants who were attending treatments, the former collected data from patient records with data for all patients, including those who had passed on. A positive aspect of this finding, or at least the current practice, is to ensure all patients are staged (classified) before the commencement of treatment. This ensures that chemotherapy regimens are selected based on disease extent and risk category, allowing clinicians to tailor drug combinations and dosing intensity appropriately. Early and accurate staging, therefore, facilitates the timely initiation of the most suitable treatment protocols and helps prevent under- or overtreatment. This is contrary to the study done in Northwest Ethiopia, in which most of the recruited pediatric cancer patients, 59.2% (n=172), had an undefined stage of cancer, followed by stage III pediatric cancer patients (20.95%).³⁰ Lack of staging may be due to a lack of or inaccessibility to diagnostic services. This may lead to the prescription and initiation of the wrong treatment protocols, which may eventually result in negative outcomes.

In the current study, a total of 21 anticancer agents were used in the treatment, mostly in a combination regimen with three drug agents as per treatment protocol requirements. Doxorubicin, vincristine, cyclophosphamide, and etoposide-based regimens were mostly prescribed, all of which are listed in the World Health Organization’s (WHO) Model List of Essential Medicines for Children (EMLc) and are in the top five medicines preferred in the LMICs.³¹ This may be attributed to the similarity of the patterns of childhood cancer between different care settings. These agents form the backbone of the treatment of childhood leukaemia and lymphoma, which are the most prevalent types of childhood cancer in Eastern Africa.

Chemotherapy-Induced Toxicities

In the current study, the most frequently reported CITs were alopecia (83%), taste changes (74%), vomiting (44%), nausea (35%), and fatigue (30%). These patterns are consistent with findings from previous pediatric oncology studies, where gastrointestinal and cutaneous toxicities are commonly observed across different populations due to similar mechanisms of cytotoxicity.^32–34 As the gastrointestinal mucosa and hair follicles contain rapidly dividing cells, they are particularly vulnerable to the effects of non–cell-specific chemotherapeutic agents. Gastrointestinal symptoms, especially vomiting, nausea, taste alterations, and loss of appetite, were among the most clinically significant toxicities because of their direct impact on nutritional intake, which may contribute to undernutrition and reduced chemotherapy tolerance. Fatigue, often related to anemia or inadequate dietary intake, may further impair daily activities and affect children’s overall quality of life. Although alopecia is an expected effect of many chemotherapy regimens, its high frequency and emotional burden highlight the importance of early counselling and psychosocial support for both patients and caregivers.

Participants experienced between one and 13 CITs each, with a mean of five toxicities per participant, aligning with reports from a Mexican pediatric oncology study in which CIT prevalence ranged widely across treatment phases.³⁵ This similarity suggests that CIT patterns among children undergoing chemotherapy are generally consistent across diverse racial and geographic settings.

Overall, the CIT profile observed shows the importance of routine monitoring, caregiver education, and timely management strategies. Early recognition and intervention, such as symptomatic treatment or dose modification, can help prevent progression to more severe toxicity, minimize treatment interruptions, and improve clinical outcomes in pediatric cancer care.

The Severity of the Reported Chemotherapy-Induced Toxicities

In the current study, most participants had CITs in grades 1 and 2, with a few in grade 3. There was no report of grade 4 (life-threatening) or grade 5 (death-related) CITs. This is because the current study included participants who were actively receiving treatments. Also, the improvement of treatment in this setting due to the implementation of best practices may have reduced the emergence of severe life-threatening CITs. On the contrary, a study at Jimma University found that most CITs were grade 3 (severe), grade 4 (life-threatening), and grade 5 (death-related).¹⁵ This could be attributed to variations in health care practices and access to medical care, which may influence the management and reporting of adverse drug reactions, but also to different patient populations with varying genetic backgrounds, medical histories, and underlying health conditions. In the current study, approximately one-third of participants did not receive any supportive medication. This may be related to the predominance of mild (grade 1) toxicities observed, for which pharmacologic intervention might not have been required. However, the specific reasons for not receiving supportive therapy, such as differences in clinical judgment, access to medicines, or symptom perception, were not assessed and remain beyond the scope of this study.

Determinants of the Reported Chemotherapy-Induced Toxicities

In the multivariable analysis, cancer classification was the only significant predictor of CITs. Children with solid tumors had approximately seven times higher odds of experiencing ≥ Grade 2 CITs compared with those with hematologic malignancies (AOR = 7.42, 95% CI 1.02–53.79, p = 0.047). Other variables, including age group, sex, vincristine-based regimen, and usenumber of chemotherapy drugs, were not significantly associated with CITs.

Although the adjusted odds ratios for older age groups appeared lower relative to children under three years, these associations were not statistically significant and were accompanied by wide confidence intervals, indicating that the observed differences may be due to chance. However, previous studies have reported age-related differences in the occurrence of CITs. For example, Gonaz Vaseghi et al found that younger children experienced higher rates of toxicity, and Abha A. Gupta et al reported that adolescents with rhabdomyosarcoma experienced fewer hematologic toxicities compared with younger patients despite receiving optimal chemotherapy doses.^36,37 These discrepancies may reflect differences in study design, sample size, cancer types included, and assessment methods across studies. Unlike these prior studies, the present work used a 7-day recall, patient- and caregiver-reported outcome tool, which may yield different sensitivity patterns in detecting toxicities across age groups. However, from a physiological point of view, young patients are more likely to suffer from CIT than adults, and chemotherapies have different potential to cause CIT.³⁸

Furthermore, the adjusted odds ratio for sex (AOR = 0.92, p = 0.893) was close to 1, suggesting no meaningful difference in toxicity risk between males and females. Factors such as differences in hormonal status and body composition in adult women, which influence pharmacokinetics and pharmacodynamics and often result in lower dose requirements compared with men, as reported by Davidson et al are not present in female children.³⁹ This may explain why, in the current study, male and female pediatric patients received similar doses and demonstrated comparable toxicity profiles. However, the report by Meeske et al, which focused on Comparative Toxicity by Sex Among Children Treated for Acute Lymphoblastic Leukaemia, showed that females presented with increased related treatment late effects than males.⁴⁰ But contrary to the current study, which has assessed CITs in a recall period of 7 days and in all Childhood cancers, their study only presents late effects and specifically in Acute Lymphoblastic leukemia.

Regimens containing vincristine were predominantly linked with peripheral neuropathy and related neurological symptoms, reflecting the known neurotoxic profile of this agent. Apart from neurological effects, vincristine generally causes less pronounced toxicities and is typically used for a short duration at the lowest effective dose. In practice, regimen selection at BMC may also be influenced by the availability of chemotherapy and supportive care medicines, as supply interruptions and resource constraints are common challenges in many low- and middle-income countries. These contextual factors may occasionally limit full implementation of standard protocols, although regimen optimization remains guided by international recommendations and national guidelines.⁴¹

Nevertheless, the majority of participants in the current study received chemotherapy regimens consisting of three anticancer agents, most of which were vincristine-based. This may partly explain why most toxicities observed were low-grade. Evidence indicates that both an increase in the number of chemotherapeutic drugs and the use of certain drug classes are associated with a higher likelihood of developing CITs. For instance, Gashaw Workalemahu et al in Ethiopia reported that pediatric oncology patients receiving four or more drugs were at increased risk of toxicity.³⁰ Similarly, Jessica Liliana Vargas Neri et al in Mexico found that anthracycline-based combination regimens were associated with at least one reported CIT.⁴² In addition, a higher number of chemotherapy cycles increases cumulative exposure, thereby raising the risk of developing CITs. In general, the observed burden of CITs likely reflects cumulative exposure to standard multi-agent regimens over several cycles, with the validated Ped-PRO-CTCAE tool enabling better detection of mild or subjective symptoms. Therefore, the safety awareness of chemotherapy use in patients with these factors calls for the need for close monitoring and surveillance of patients to improve the outcome of treatment and minimise toxicity effects.⁴³

In Tanzania, ADR reporting is coordinated by the Tanzania Medicines and Medical Devices Authority (TMDA), which serves as the national pharmacovigilance center and contributes to the global database managed by the WHO Collaborating Center for International Drug Monitoring (Uppsala Monitoring Center). However, CITs, particularly in pediatric populations, remain underreported in this system.⁴⁴ The findings of the present study therefore, complement the limited national pharmacovigilance data by providing more detailed, patient-level insights into the frequency and patterns of chemotherapy-related toxicities among children.

Study Limitations

This study has several limitations. The cross-sectional design precludes causal inference, and the convenience sampling from a single center may limit generalizability to other settings. Because only patients who were actively attending the oncology clinic were eligible for recruitment, children who were critically ill or unable to present for follow-up were not captured. This selection bias may partly explain the predominance of grade 1–2 toxicities and the absence of grade 4–5 events. Reliance on child- and caregiver-reported data may also introduce recall or reporting bias, particularly among younger children or for non-visible symptoms. Although the Ped-PRO-CTCAE is a validated tool, self-reported severity may differ from clinician-based grading, and the absence of laboratory confirmation for some toxicities may affect diagnostic precision.

Conclusion

A total of 43 distinct types of CITs were identified, comprising 692 symptom-level events reported across all participants. Although individual children experienced multiple toxicities, the majority were mild (grade 1), and only a smaller proportion reached grade 2 or higher. Gastrointestinal toxicities, such as taste changes, vomiting, nausea, and loss of appetite, and skin-related effects, including alopecia, itchy red bumps, and dry skin, were the most frequently reported. Moreover, cancer classification was the only factor significantly associated with experiencing ≥ Grade 2 toxicity, with children diagnosed with solid tumors showing higher odds of moderate-to-severe CITs compared with those with hematologic malignancies. Routine adoption of structured patient- and caregiver-reported outcome tools may strengthen early toxicity detection, enhance supportive care, and improve treatment tolerance among children receiving chemotherapy.