Treatment outcomes of pediatric acute myeloid leukemia in Western Kenya before and after the implementation of the SIOP PODC treatment guideline

Abstract Purpose The Pediatric Oncology in Developing Countries (PODC) committee of the International Society of Pediatric Oncology (SIOP) published a pediatric acute myeloid leukemia (AML)‐specific adapted treatment guideline for low‐ and middle‐income countries. We evaluated the outcomes of children with AML at a large Kenyan academic hospital before (period 1) and after (period 2) implementing this guideline. Patients and Methods Records of children (≤17 years) newly diagnosed with AML between 2010 and 2021 were retrospectively studied. In period 1, induction therapy comprised two courses with doxorubicin and cytarabine, and consolidation comprised two courses with etoposide and cytarabine. In period 2, a prephase with intravenous low‐dose etoposide was administered prior to induction therapy, induction course I was intensified, and consolidation was adapted to two high‐dose cytarabine courses. Probabilities of event‐free survival (pEFS) and overall survival (pOS) were estimated using the Kaplan–Meier method. Results One‐hundred twenty‐two children with AML were included – 83 in period 1 and 39 in period 2. Overall, 95 patients received chemotherapy. The abandonment rate was 19% (16/83) in period 1 and 3% (1/39) in period 2. The early death, treatment‐related mortality, complete remission, and relapse rates in periods 1 and 2 were 46% (29/63) versus 44% (14/32), 36% (12/33) versus 47% (8/17), 33% (21/63) versus 38% (12/32), and 57% (12/21) versus 17% (2/12), respectively. The 2‐year pEFS and pOS in periods 1 and 2 were 5% versus 15% (p = .53), and 8% versus 16% (p = .93), respectively. Conclusion The implementation of the SIOP PODC guideline did not result in improved outcomes of Kenyan children with AML. Survival of these children remains dismal, mainly attributable to early mortality.


| INTRODUCTION
Annually, there are approximately 429 000 new childhood cancer cases worldwide, but almost half are undiagnosed. [1][2][3][4][5] One-third of these new cases are leukemia, of which 15%-20% concern acute myeloid leukemia (AML). Since over 80% of all new cases occur in low-and middle-income countries (LMICs), 1 around 20 000 children develop AML in these countries yearly. 6 In high-income countries (HICs), long-term probabilities of overall survival (pOS) of children with AML currently exceed 70%. [7][8][9][10] Considerably poorer rates have been reported in LMICs, 11 in particular in sub-Saharan Africa. [12][13][14] Together with low salvage rates after relapse, high rates of treatment abandonment, early death (ED), and treatmentrelated mortality (TRM) mainly explain these inferior outcomes in LMICs. 11 Furthermore, limited resources, such as the unavailability of chemotherapeutics and essential supportive care measures (e.g., antibiotics, blood products), add to the poor outcomes. 11 Consequently, the management of pediatric AML in LMICs is challenging.
In 2019, the Pediatric Oncology in Developing Countries (PODC) committee of the International Society of Pediatric Oncology (SIOP) published a pediatric AML-specific adapted treatment guideline for level 2 settings in LMICs as per PODC framework definition. 6,15 Recommendations in this guideline include a prephase with low-dose chemotherapy prior to standard treatment, either with oral etoposide combined with 6-thioguanine and prednisolone (PrET), or with singleagent etoposide administered orally or intravenously (i.v.), and a highdose cytarabine (HDAC) consolidation regimen on three consecutive days, instead of on days 1, 3, and 5. A prephase would put the leukemia in partial remission and would provide time to treat any active infections, improve the patients' clinical condition and nutritional status, and arrange financial support. It is anticipated that with this approach, the following induction therapy can be administered more safely, hopefully resulting in lower TRM and subsequent improved OS. A 3-day consolidation course would result in shortened hematologic recovery time, lower need for platelet transfusions, and less infections, without decreasing survival. 6 The Moi Teaching and Referral Hospital (MTRH) in Eldoret, Western Kenya, implemented the SIOP PODC treatment guideline. The aim of this study was to evaluate the outcomes of Kenyan children with AML treated before and after this implementation.

| Setting
Kenya is a sub-Saharan lower-middle-income country in East Africa with a population of approximately 55 million. 16 A government-owned health-insurance system is available, the National Hospital Insurance Fund (NHIF), which, after enrollment and activation, covers inpatient treatment costs in public health care facilities. 17,18 However, only 15% of Kenyans have NHIF and the majority pays health care out-ofpocket. 17 The Shoe4Africa Children's Hospital at MTRH is one of the few public hospitals where pediatric oncology patients can be treated comprehensively. This hospital is a level 2 setting based on infrastructural and service availability characteristics as per PODC framework definition. 6 Note: During study period 1, the infusion rate of doxorubicin changed from 1 to 4 h, the dose of etoposide was increased from 100 to 200 mg/m 2 , and the dose of cytarabine in the consolidation phase was lowered from 200 to 100 mg/m 2 . During both study periods, patients were started on allopurinol 10 mg/kg thrice daily 24 h prior to the start of chemotherapy. As long as tumor load was high, patients were hyperhydrated with 3000 mL/m 2 24 h before and after chemotherapy administration. Peripheral blood count recovery criteria were an ANC >1.0 Â 10 9 /L, platelets >150 Â 10 9 /L, and WBC >2.0 Â 10 9 /L during study period 1 and an ANC >1.0 Â 10 9 /L and platelets >75 Â 10 9 /L during study period 2. Refractory and relapsed patients were referred to palliative care. Hematopoietic stem cell transplantation was not available. guideline, referred to as study period 1 (January 2010 -July 2019), and after its implementation, referred to as study period 2 (August 2019 -December 2021). The main treatment differences between the two periods were the administration of a prephase with i.v. low-dose etoposide prior to standard treatment, an intensified induction course I, a HDAC consolidation regimen, and the expansion of supportive care measures in period 2. Intravenous etoposide as prephase regimen was chosen over PrET and oral etoposide because of its greater availability. During both periods, blood cultures were not routinely taken due to long processing times and frequent unavailability of blood culture bottles.

| Definitions and response criteria
Treatment abandonment was defined as failure to initiate or continue treatment with curative intent during the first four consecutive weeks after registration in the hospital. 21 Complete remission (CR) was defined as <5% leukemic blasts by morphology in a bone marrow aspirate, assessed prior to consolidation therapy, with clear evidence of hematological regeneration in the bone marrow and in the peripheral blood.
Refractory disease was defined as the failure to achieve CR. ED was defined as death within 42 days of treatment. TRM was defined as any death occurring after 42 days of treatment, in the absence of progressive cancer. 22 Relapse diagnosis was based on the reappearance of leukemic blasts in the peripheral blood, or the occurrence of extramedullary disease after initial CR. Event-free survival (EFS) was defined as the time from diagnosis until the first event, or last follow-up. Events included induction failure (i.e., an immediate palliative care decision after the medical team's opinion on incurability, death prior to initiation of intended treatment, ED, death occurring after 42 days of treatment but prior to CR assessment, or refractory disease), treatment abandonment, death in CR, and relapse. OS was defined as the time from diagnosis until death, or last follow-up.    The median diagnosis delays were 7 (IQR, 3-13) and 5 (IQR, 3-9) days, and the median treatment delays were 16 (IQR, 9.8-28.3) and 7 (IQR, 3-12.8) days in periods 1 and 2, respectively.    Figure 2).

| DISCUSSION
This study evaluated outcomes of Kenyan children with AML before and after implementing the SIOP PODC pediatric AML-specific adapted treatment guideline. There was no significant difference in the 2-year pEFS (5% vs. 15%; p = .53) and pOS (8% vs. 16%; p = .93) after the implementation of the guideline.
The prephase was recommended to lower TRM and improve OS. However, the TRM rate was 47% in period 2 and 36% in period 1, and the 2-year pOS was not statistically significantly different (16% vs. 8% in period 1). Our results cannot be compared with those achieved by others as, to the best of our knowledge, no previous studies evaluated the efficacy of the guideline.
ED rates were remarkably high in both periods. Induction course I included intensified cytarabine during period 2, thereby possibly negating the potential benefit of the prephase. It may be considered to administer standard two or three repetitive prephase cycles, since all patients in period 2 except one received only one prephase cycle, or to administer a low-dose induction course I. The aim hereof is to lower toxicity. These alternatives are also described in the SIOP PODC guideline. 6 Our findings highlight the challenging balance between anticancer treatment and supportive care measures, as a structured approach of supportive care is key to reduce morbidity and mortality in these resource-limited settings. In LMICs, bacterial and fungal infections during neutropenic episodes are the main cause of mortality. 23 Despite the expansion of antibacterial prophylaxis and the increased availability of antibiotics in our setting, many patients died of sepsis. It has to be acknowledged that sepsis was most often a clinical diagno- tures should be studied in order to gain more insight into the causative pathogens and resistance patterns, so that targeted antibiotics can be initiated promptly. Education of the parents and nurses on the importance of nasogastric feeding will also be an important aspect.
Additionally, future focus should be on the optimization of prephase and induction therapy. This may include prolongation of the prephase and/or administering a low-dose induction course I. Currently, there are plans to build the first pediatric oncology hospital in sub-Saharan Africa at MTRH with a 100-bed capacity. Hopefully, all these efforts will ultimately result in an improved survival of children with AML in sub-Saharan Africa.

CONFLICT OF INTEREST STATEMENT
The authors have no conflict of interest to disclose.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.

ETHICS STATEMENT
We declare that the work submitted to Cancer Reports has been done in accordance to the COPE principles, and that it has been performed in an ethical and responsible way, with no research misconduct, which includes, but is not limited to data fabrication and falsification, plagia-