An AI based, open access screening tool for early diagnosis of Burkitt lymphoma.

Burkitt Lymphoma (BL) is a highly treatable cancer. However, delayed diagnosis of BL contributes to high mortality in BL endemic regions of Africa. Lack of enough pathologists in the region is a major reason for delayed diagnosis. The work described in this paper is a proof-of-concept study to develop a targeted, open access AI tool for screening of histopathology slides in suspected BL cases. Slides were obtained from a total of 90 BL patients. 70 Tonsillectomy samples were used as controls. We fine-tuned 6 pre-trained models and evaluated the performance of all 6 models across different configurations. An ensemble-based consensus approach ensured a balanced and robust classification. The tool applies novel features to BL diagnosis including use of multiple image magnifications, thus enabling use of different magnifications of images based on the microscope/scanner available in remote clinics, composite scoring of multiple models and utilizing MIL with weak labeling and image augmentation, enabling use of relatively low sample size to achieve good performance on the inference set. The open access model allows free access to the AI tool from anywhere with an internet connection. The ultimate aim of this work is making pathology services accessible, efficient and timely in remote clinics in regions where BL is endemic. New generation of low-cost slide scanners/microscopes is expected to make slide images available immediately for the AI tool for screening and thus accelerate diagnosis by pathologists available locally or online.

Expanding Myeloma Training and Care in Western Kenya Through the ECHO Model: The Pilot Phase.

PURPOSE: Multiple myeloma (MM) in rural western Kenya is characterized by under and late diagnosis with poor long-term outcomes. Inadequate skilled rural health care teams are partly to blame. The Extension for Community Healthcare Outcomes (ECHO) model attempts to bridge this skills gap by linking rural primary/secondary health care teams (spokes) to myeloma experts in a tertiary care center (hub) in a longitudinal training program. METHODS: A hub team comprising myeloma experts and administrators from Moi Teaching and Referral Hospital/Academic Model Providing Access to Healthcare was assembled and spoke sites were recruited from rural health care facilities across western Kenya. A curriculum was developed by incorporating input from spokes on their perceived skills gaps in myeloma. Participants joined sessions remotely through virtual meeting technology. ECHO sessions consisted of a spoke-led case presentation with guided discussion followed by an expert-led lecture. An end-of-program survey was used to evaluate participant satisfaction, knowledge, and practice patterns. RESULTS: A total of eight sessions were conducted between April and November 2021 with a median of 40 attendees per session drawn from diverse health care disciplines. Twenty-four spoke sites were identified from 15 counties across western Kenya. The majority of attendees reported satisfaction with the ECHO program (25 of 29) and improvement in their myeloma knowledge (24 of 29). There were 74 new myeloma diagnoses made at the hub site in 2021, representing a 35% increase from the previous 3-year average despite the COVID-19 pandemic that suppressed health care access globally. RECOMMENDATIONS: The pilot ECHO model was successfully implemented in myeloma training for rural-based health care teams. Key attributes included collaborative curriculum development, interactive case-based bidirectional learning, and multidisciplinary engagement.

Improvement of diagnosis in children with Burkitt lymphoma in Kenya: feasibility study for the implementation of fluorescence in situ hybridisation testing for MYC and the MYC/IGH translocation.

Background: Indiana University (IU) initiated fluorescence in situ hybridisation (FISH) methodology for Burkitt Lymphoma (BL) to advance the accuracy and speed of diagnosis in the AMPATH Reference Laboratory at Moi Teaching and Referral Hospital (MTRH) in Eldoret, Kenya. Standard diagnostic testing for BL at MTRH includes morphology of the biopsy specimen or aspirate and limited immunohistochemistry panels. Methods: Tumour specimens from 19 children enrolled from 2016 to 2018 in a prospective study to improve the diagnosis and staging of children with suspected BL were evaluated. Touch preps from biopsy specimens or smears from fine needle aspiration were collected, stained with Giemsa and/or H&E and reviewed by pathologists to render a provisional diagnosis. Unstained slides were stored and later processed for FISH. Duplicate slides were split between two laboratories for analysis. Flow cytometry results were available for all specimens. Results from the newly established FISH laboratory in Eldoret, Kenya were cross-validated in Indianapolis, Indiana. Results: Concordance studies found 18 of 19 (95%) of specimens studied yielded analysable FISH results for one or both probe sets (MYC and MYC/IGH) in both locations. There was 94% (17/18) concordance of results between the two FISH laboratories. FISH results were 100% concordant for the 16 specimens with a histopathological diagnosis of BL and two of three non-BL cases (one case no result in IU FISH lab). FISH was similarly concordant with flow cytometry for specimens with positive flow results with the exception of a nasopharyngeal tumour with positive flow results for CD10 and CD20 but was negative by FISH. The modal turn-around time for FISH testing on retrospective study specimens performed in Kenya ranged between 24 and 72 hours. Conclusion: FISH testing was established, and a pilot study performed, to assess the feasibility of FISH as a diagnostic tool for the determination of BL in a Kenyan paediatric population. This study supports FISH in limited resource settings to improve the accuracy and speed of diagnosis of BL in Africa.

Retrospective Analysis of Presentation, Treatment, and Outcomes of Multiple Myeloma at a Large Public Referral Hospital in Eldoret, Kenya.

PURPOSE: Treatment patterns and survival outcomes of patients with multiple myeloma (MM) in Kenya have not been adequately characterized. The objectives of this study were to describe the clinical, laboratory, and imaging findings at diagnosis, to describe the treatment offered, and to determine the survival outcomes of patients with MM over an 11-year period. PATIENTS AND METHODS: A retrospective chart review was carried out for all patients who were diagnosed and treated for MM at Moi Teaching and Referral Hospital from 2009 to 2019. The Kaplan-Meier method was used to estimate survival. Factors affecting survival were identified using univariate and multivariate analyses. RESULTS: A total of 221 patient charts were analyzed of which 124 belonged to male patients (56.1%). The median age at diagnosis was 61 years. Bone pain was the most common presenting complaint observed in 69.6% of 194 patients assessed. Out of 102 patients who received imaging studies, 60 (58.8%) had lytic lesions, 30 (29.4%) had fractures, whereas 30 (29.4%) had spinal cord compression. Anemia, renal failure, and hypercalcemia were observed in 87/187 (46.5%), 22/161 (13.7%), and 23/42 (54.8%) patients, respectively. Thalidomide and dexamethasone (65.2%); bortezomib, thalidomide, and dexamethasone (14.6%); and melphalan and prednisolone (11.9%) were the most prescribed initial chemotherapy regimens among 219 patients analyzed. Overall survival at 1 and 5 years was 70% and 21%, respectively; median overall survival was 29.0 months. In multivariate analysis, male sex (hazard ratio [HR] 1.9), baseline anemia (HR 1.8), and baseline renal failure (HR 3.2) were associated with significantly shorter survival. CONCLUSION: Survival outcomes were poor despite increased use of multiagent-based chemotherapy regimens. Greater access to available diagnostics and treatments is required to achieve rational treatment and increased survival.

Building a Sustainable Comprehensive Multiple Myeloma Program in Western Kenya.

Despite improved treatment strategies for multiple myeloma (MM), patient outcomes in low- and middle-income countries remain poor, unlike high-income countries. Scarcity of specialized human resources and diagnostic, treatment, and survivorship infrastructure are some of the barriers that patients with MM, clinicians, and policymakers have to overcome in the former setting. To improve outcomes of patients with MM in Western Kenya, the Academic Model Providing Access to Healthcare (AMPATH) MM Program was set up in 2012. In this article, the program's activities, challenges, and future plans are described distilling important lessons that can be replicated in similar settings. Through the program, training on diagnosis and treatment of MM was offered to healthcare professionals from 35 peripheral health facilities across Western Kenya in 2018 and 2019. Access to antimyeloma drugs including novel agents was secured, and pharmacovigilance systems were developed. Finally, patients were supported to obtain health insurance in addition to receiving peer support through participation in support group meetings. This article provides an implementation blueprint for similar initiatives aimed at increasing access to care for patients with MM in underserved areas.

Towards a unified approach for multiple myeloma care in Kenya - proceedings of the Inaugural Multiple Myeloma Congress.

The rising burden of multiple myeloma in Kenya has not been met by a commensurate effort for control. Patients and practitioners struggle with unavailability and unaffordability of diagnostics, drugs and stem cell transplant leading to presentation at advanced stages and under-treatment with increased morbidities and mortality. A concerted effort among stakeholders is urgently needed to develop strategies for myeloma control. The scarcity of providers also carries grave consequences for Kenyan patients. The Academic Model Providing Access To Healthcare (AMPATH) multiple myeloma program organized the Inaugural Virtual Multiple Myeloma Congress to achieve both interactive specialist instruction and stakeholder engagement. Expert presenters and panellists from diverse disciplines were invited to offer in-depth presentations on myeloma care and case studies from panellists´ practice were used to contextualize learning points and form a basis for generating debate on the challenges facing providers and opportunities for care improvement. An audience of health professionals offering care to myeloma patients was invited. The underlying principle of recommendations developed during the congress was collaboration among in-country and international practitioners, researchers and policy experts from private and public sector. This partnership of stakeholders bears the potential of pooling scarce resources and for collective advocacy towards better patient care.