Despite the fact that simple interventions and emerging therapies can increase survival of SCD patients, under detection renders SCD a major cause of under age-5 mortality in rural regions of the developing world where it is prevalent. (Piel et PLoS Med DOI: org/10.1371/journal.pmed.1001484, 2013). Two major diagnostic challenges are: 1. A lack of affordable diagnostic methodology to obtain point-of-care results in low-resource environments. 2. Dispersion of SCD patients in vast rural hinterlands who rarely interface with healthcare services.
To address the first impediment, we have developed and validated a simple, low-cost test to screen for SCD using aqueous multiphase systems (AMPS). AMPS are solutions of polymers and surfactants that form immiscible self-assembling step-gradients. An AMPS designed to target the density signature of sickle cell anemia forms the basis of this novel diagnostic technique. (Kumar et al, PNAS DOI:10.1073/pnas.14147391110, 2014). Building on a design previously tested in Zambia (Kumar et al, PLoS One DOI: 10.1371/journal.pone.0114540, 2014) an improved test has been developed and named "Mpana" for "Multi-Phase Analyzer." The Mpana test detects the presence or absence of cells with a high mass-density as an indirect method to test for the presence of sickled cells. The test uses 5 µL of fingerstick blood, can be performed in 15 minutes using an inexpensive ($150), battery-powered centrifuge, the results can be read with the naked eyes by a trained reader (Figure 1) and costs 50 cents per test. The improved density-based test was first verified on fingerstick samples in the U.S. before being validated in a field study in rural Zambia.
To manage the access difficulty of large-scale rural SCD screening, we exploited a partnership with the National Zambian Dental Training School. We have previously demonstrated that provision of free toothbrushes, toothpaste and other minimal health services is a powerful magnet to attract subjects in rural regions of Zambia that have high rates of tooth decay. Once collected these individuals can receive other healthcare interventions such as SCD screening.
We conducted a cross-sectional field test of this approach in August and September of 2017 in the rural Serenje and Chitambo Districts of Northern Zambia. A local drama group and radio and mobile broadcasts informed the target communities about the dental intervention that attracted over 700 subjects. Of these, 503 children aged 1 month to 18 years were symptom screened for SCD, tested with MPANA, and dried blood spot (DBS) samples collected. The DBS were analyzed by iso-electric focusing and high-performance liquid chromatography at the New England Newborn Screening Program. Of the screened subjects, 78 had sickle cell trait and 17 had SCD. A score was assigned to visual readings of the MPANA test to enable ROC analysis.The ROC curve shows good diagnostic performance with an area under the curve (AUC) of 0.84. (Figure 2) The MPANA test had a sensitivity of 71% (CI 47-88%) and a specificity of 99% (CI 98-100%), with an overall diagnostic accuracy of 98% (CI 97-99%). Importantly, the MPANA test had a negative predictive value of 99% (CI 98-100%). Combining Mpana results with a clinical symptom score yielded a slightly improved specificity of 100% (CI 99.5-100%) and yielded an overall diagnostic accuracy of 99% (CI 98-100%).
In conclusion, a multidisciplinary consortium has achieved the first validation of a low-cost simple point of care test to diagnose SCD in a rural area of a developing nation and has demonstrated the ability of dental services to facilitate such screening. Offsetting the theoretical concern that the MPANA test might not detect SCD in neonates and young infants due to their high hemoglobin F content are the advantages that our SCD screening strategy is low cost, primarily utilizes local resources and immediately delivers positive subjects to care interventions (which could be anything in addition to or other than SCD screening). It is therefore amenable to scaling up for widespread application.
Acknowledgments: The Provincial and District Offices, Rural Health Centers and communities that granted permissions and participated in this study.
Author Contributions: Contributed equally to the work - CMCL,AAK,TPS
Conflict-of-interest disclosure: Competing financial interests declared.
Correspondence: Catherine Chunda-Liyoka, e-mail: firstname.lastname@example.org
Kumar:Harvard University: Patents & Royalties: I am an inventor on patents for the underlying technology and the patent has been assigned to Harvard.. Humpton:Nano Terra, Inc.: Employment. Whitesides:Harvard University: Patents & Royalties: Harvard holds patents for this work. Stossel:Harvard University: Patents & Royalties: Co-inventor on MPANA patent. No royalties.
Asterisk with author names denotes non-ASH members.