Evaluation of Low-cost Techniques for Detecting Sickle Cell Disease and β-thalassemia in Nepal and Canada

Sickle cell disease (SCD) is an inherited blood disorder associated with acute illness and organ damage. In high resource settings, early screening and treatment greatly improve quality of life. In low resource settings, however, mortality rate for children is high (50-90%). Low-cost and accurate screening techniques are critical to reducing the burden of the disease, especially in remote/rural settings. The most common and severe form of SCD is sickle cell anemia (SCA), caused by the inheritance of genes causing abnormal forms of hemoglobin (called sickle hemoglobin or hemoglobin S) from both parents. The asymptomatic or carrier form of the disease, known as sickle cell trait (SCT), is caused by the inheritance of only one variant gene from one of the parents. In areas such as Nepal, β-thalassemia (another inherited blood disorder) and SCD are both prevalent, and some combinations of these diseases lead to severe symptoms.

The purpose of this study is to determine the accuracy of low-cost point-of-care techniques for screening and detecting sickle cell disease, sickle cell trait, and β-thalassaemia, which will subsequently inform on feasible solutions for detecting the disease in rural, remote, or low-resource settings. One of the goals of the study is to evaluate the feasibility of techniques, such as the sickling test with low-cost microscopy and machine learning, HbS solubility test, commercial lateral-flow assays (HemoTypeSC and Sickle SCAN), and the Gazelle Hb variant test, to supplement or replace gold standard tests (HPLC or electrophoresis), which are expensive, require highly trained personnel, and are not easily accessible in remote/rural settings.

The investigators hypothesize that:

1. an automated sickling test (standard sickling test enhanced using low-cost microscopy and machine learning) has a higher overall accuracy than conventional screening techniques (solubility and sickling tests) to detect hemoglobin S in blood samples
2. the automated sickling test can additionally classify SCD, SCT and healthy individuals with a sensitivity greater than 90%, based on morphology changes of red blood cells, unlike conventional sickling or solubility tests that do not distinguish between SCD and SCT cases
3. Gazelle diagnostic device can detect β-thalassaemia and SCD/SCT with an overall accuracy greater than 90%, compared with HPLC as the reference test

Study Overview

• Status: Completed
• Conditions: Sickle Cell Disease; Beta-Thalassemia; Sickle Cell Trait; Sickle Cell-Beta Thalassemia; Sickle Cell-SS Disease
• Intervention / Treatment: Diagnostic test: High performance liquid chromatography; Device: Automated sickling test; Diagnostic test: HbS solubility test; Device: HemoTypeSC; Device: Sickle SCAN; Device: Gazelle Hb Variant Test

Detailed Description

Overall, the hypothesis is that an assessment of the performance and accuracies of low-cost point-of-care techniques (automated sickling test, solubility test, lateral-flow assays, Gazelle Hb variant test) against HPLC tests will provide researchers and health workers with feasible alternative options for screening and detecting SCD, SCT and β-thalassaemia in a variety of situations based on the needs of the communities and the resources available.

Objectives

Objectives specific to the current study are to:

1. Determine accuracy (sensitivity and specificity) of automated sickling test to detect HbS, compared to gold standard HPLC, and to conventional solubility test
2. Determine whether SCD, SCT and healthy individuals can be classified using the automated sickling test that leverages machine learning on images of blood films under hypoxia
3. Validate accuracy (>95% sensitivity and specificity) of lateral- flow assays (HemoTypeSC and Sickle SCAN) to detect SCD/SCT, and of Gazelle variant test to detect SCD, SCT, and β-thalassaemia; and determine if low-cost techniques can potentially replace HPLC/electrophoresis tests in rural and remote settings

Long-term objectives of the overall project are to:

1. Implement trained machine learning algorithm to classify SCD, SCT and healthy individuals during screening tests in Nepal
2. Implement relevant low-cost point-of-care techniques in rural and remote communities of Nepal using insights and conclusions from current study

The plan of the study to screen the communities (e.g. in Nepalgunj, in Vancouver) using the following:

a. Low-cost screening i. Sickling test with low-cost microscope and automated screening with machine learning ii. Sickling test with traditional microscope (conventional manual screening used in Nepal) iii. HbS solubility test iv. Commercial point-of-care assays (HemoTypeSC and Sickle SCAN) v. Gazelle Hb variant test b. Gold standard test: HPLC, for determining the accuracies of low-cost screening techniques

De-identified data (images of blood films and associated documentation) will also be deposited in an online public repository, such as the Federated Research Data Repository (FRDR). FRDR is a service of the Digital Research Alliance of Canada (Alliance), a not-for-profit organization that supports digital research infrastructure in Canada. FRDR is hosted on national infrastructure, managed and administered by the Digital Research Alliance of Canada.

• Study Type: Interventional
• Enrollment (Actual): 138
• Phase: Not Applicable

Contacts and Locations

• Study Contact: Name: Pranav Shrestha; Phone Number: 2508583404; Email: pranav.shrestha@ubc.ca
• Study Contact Backup: Name: Boris Stoeber; Phone Number: 6048275907; Email: boris.stoeber@ubc.ca

Study Locations

• Canada
  • British Columbia
    • Vancouver, British Columbia, Canada, V6Z 1Y6
      • St. Paul's Hospital
    • Vancouver, British Columbia, Canada, V6H 3N1
      • BC Children's Hospital
• Nepal
  • Banke
    • Nepalgunj, Banke, Nepal
      • Mount Sagarmatha Polyclinic and Diagnostic Center

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 1 year and older (Child, Adult, Older Adult)
• Accepts Healthy Volunteers: Yes

Description

Inclusion Criteria:

Since the techniques evaluated in the study aims at detecting sickle cell disease (SCD), sickle cell trait (SCT), and β- thalassemia, the following number of participants will be included in Nepal:

• 20 individuals with SCD (HbSS)
• 20 individuals with SCT (HbAS)
• 20 individuals with sickle cell/β-thalassemia compound heterozygous form (HbS/β-thalassemia)
• 20 individuals with β-thalassemia (Hbβ/β-thalassemia)
• 20 individuals with β-thalassemia trait or carrier form (HbA/β- thalassemia)
• 20 healthy individual participants or normal participants (HbAA, participants without any known hemoglobin disorders, such as SCD, SCT or β-thalassemia)

The following number of participants will be included in Canada:

• 30 individuals with SCD (HbSS)
• 30 individuals with SCT (HbAS)
• 30 healthy individual participants or normal participants (HbAA, participants without any known hemoglobin disorders, such as SCD, SCT or β-thalassemia)

Participants older than 1 year of age at the time of drawing blood will be eligible. Signed and dated consent or assent forms will be required by the participants or their parents/guardians.

Exclusion Criteria:

The exclusion criteria for the study:

• Transfusion within the last 3 months
• Pregnancy Participants who wish to withdraw from the study will also be excluded.

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Diagnostic
• Allocation: N/A
• Interventional Model: Single Group Assignment
• Masking: None (Open Label)

Number of Arms

1

Arms and Interventions

Participant Group / Arm

Intervention / Treatment

Other: 1) HbSS; 2) HbAS; 3) HbS/β-thalassemia; 4)Hbβ/β-thalassemia; 5) HbA/β- thalassemia; 6) HbAA; Around 20 participants each (in Nepal): with the homozygous form of sickle cell disease (HbSS); with the heterozygous form of sickle cell disease (HbAS); with the compound heterozygous form of sickle cell disease (HbS/β-thalassemia); with the carrier form of β-thalassemia (HbA/β-thalassemia); with the carrier form of β-thalassemia (HbA/β-thalassemia); without any known hemoglobin disorders, such as sickle cell disease, sickle cell trait, β-thalassemia, etc. Around 30 participants each (in Canada): with the homozygous form of sickle cell disease (HbSS); with the heterozygous form of sickle cell disease (HbAS); without any known hemoglobin disorders, such as sickle cell disease, sickle cell trait, β-thalassemia, etc.

Diagnostic test: High performance liquid chromatography; High performance liquid chromatography (HPLC) using the D10 System by Bio-Rad Laboratories will be used as the gold standard test.; Device: Automated sickling test; The standard sickling test using 2% sodium metabisulphite will be augmented using an automated microscope (such as Octopi) and machine learning, and will be used as one of the low-cost tests.; Diagnostic test: HbS solubility test; Standard HbS solubility test currently used in Nepal (e.g. Sicklevue) will be used as one of the low-cost tests; Device: HemoTypeSC; A point-of-care lateral flow assay, HemoTypeSC (https://www.hemotype.com/), will be used as one of the low-cost tests; Device: Sickle SCAN; A point-of-care lateral flow assay, Sickle SCAN (https://www.biomedomics.com/products/hematology/sicklescan/), will be used as one of the low-cost tests; Device: Gazelle Hb Variant Test; A portable electrophoresis machine, Gazelle diagnostic device (https://hemexhealth.com/), will be used as one of the low-cost tests

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Sensitivity, specificity, positive predictive value and negative predictive value	The following metrics will be determined for the low-cost tests to be evaluated as indicated below (where TP = true positive, TN = true negative, FP = false positive, FN = false negative): Sensitivity = TP/(TP + FN); Specificity = TN/(FP + TN); Positive predictive value = TP/(TP + FP); Negative predictive value = TN/(TN + FN) These metrics will be calculated for the low-cost technologies against the reference test, HPLC, for detecting the presence of sickle hemoglobin and β- thalassemia. The low-cost technologies include automated sickling test (standard sickling test enhanced using low-cost microscopy and machine learning), solubility test, HemoTypeSC, Sickle SCAN, and Gazelle Hb Variant test. The test results of the low-cost technologies will be compared with those of the reference test to get the values of TP, TN, FP and FN, which will then be used to calculate the metrics listed above.	baseline

Collaborators and Investigators

• Sponsor: University of British Columbia
• Investigators: Principal Investigator: Boris Stoeber, University of British Columbia

Study record dates

Study Major Dates

• Study Start (Actual): September 20, 2022
• Primary Completion (Actual): March 30, 2023
• Study Completion (Actual): March 30, 2023

Study Registration Dates

• First Submitted: August 15, 2022
• First Submitted That Met QC Criteria: August 17, 2022
• First Posted (Actual): August 18, 2022

Study Record Updates

• Last Update Posted (Actual): August 2, 2023
• Last Update Submitted That Met QC Criteria: August 1, 2023
• Last Verified: August 1, 2023

More Information

Terms related to this study

• Keywords: Machine learning; low-cost test; point-of-care diagnostics; automated sickling test; solubility test; HemoTypeSC; Sickle Scan; Gazelle Hb Variant Test; Image database; Red blood cells
• Additional Relevant MeSH Terms: Hematologic Diseases; Genetic Diseases, Inborn; Anemia; Anemia, Hemolytic, Congenital; Anemia, Hemolytic; Hemoglobinopathies; Anemia, Sickle Cell; Thalassemia; beta-Thalassemia; Sickle Cell Trait
• Other Study ID Numbers: H22-00294 (Other Identifier: UBC Clinical Research Ethics Board); 85/2022 (Other Identifier: Nepal Health Research Council); H21-01929 (Other Identifier: UBC-Providence Health Care Research Institute)

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: YES
• IPD Plan Description: Only de-identified data and test results will be shared. The test results for the low-cost tests and HPLC tests will be published in aggregate form. De-identified images of blood films will be deposited in an online public repository, such as Federated Research Data Repository (FRDR).

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: Yes
• product manufactured in and exported from the U.S.: Yes