To validate the Aikili technology, we intend to evaluate the performance of the Aikili system against current clinical workflows for breast cancer diagnostic profiling ( immunohistochemistry). Forty (40) clinical samples will be collected by Aga Khan University Hospital, Nairobi.This study will compare the sensitivity and specificity of the Aikili diagnostic against accepted gold standards. Our goal is to achieve >90% congruency to match the sensitivity and specificity of current workflows.

One of the biggest cancer challenges in low- and middle-income countries (LMICs) is providing rapid, affordable diagnostics that enable patients to obtain locally available treatments most likely to enable survival based on personalized medicine. Delayed cancer diagnosis – delays between incidence of cancer and patients’ ability to access the pathology services that enable oncologists to prescribe treatment – occurs in many LMICs due to bottlenecks in specimen acquisition, complex specimen handling logistics, lack of pathologists per capita, and infrastructure for tertiary laboratories, which is primarily available in major urban areas. Consequently, this results in missed treatment options for many patients and increased mortality rates. For breast cancer, the most common invasive malignancy amongst women, LMICs represented the majority of >2M new cases globally in 2018 and carried a disproportionate burden of breast cancer morbidity and mortality, with 69% of quality years of life lost due to breast cancer occurring in LMICs. As such, it is of critical public health importance to develop new methods to diagnose cancer that overcome these barriers and extend quality years of life for patients.

This study seeks to enable same-day diagnostic profiling of breast cancer at the point-of-care using a low-cost, automated system. The goal of this study is to build and validate LMIC- ready diagnostic systems for the analysis of breast fine needle aspirate (FNA) samples of palpable mass lesions. FNA samples can be collected by health workers in decentralized clinics, with deeper lesions sampled using handheld ultrasound devices. The cellular specimen are processed with a fast turn-around time of <1 hour from sample to diagnosis. The end-result is an integrated platform with automated analyses that can be used by non-expert healthcare providers to process samples and escalate results to reviewing pathologists. It answers the questions on whether there are malignant cells in an FNA and the receptor status (ER/PR, HER2) of these cells.

To validate the Aikili technology, we intend to evaluate the performance of the updated Aikili system against gold standard workflows at AKU’s accredited Pathology facilities. Using 40 clinical samples collected by Aga Khan University Hospital, Kenya, this study will compare the sensitivity and specificity of the Aikili diagnostic against accepted gold standards. Our goal is to achieve >90% congruency to match the sensitivity and specificity of current workflows. Phase I of the project will be considered successful when we can show that the field-optimized Aikili system accurately and reliably detects breast cancer receptor status in human FNAs compared to accepted gold standards.