From blueprint to biobank: Leveraging expert recommendations for implementing change (ERIC) to pediatric cancer biobanking in Pakistan

Abstract

Background: In low- and middle-income countries, limited infrastructure and resources hinder biobank establishment, affecting specimen diversity. Addressing this gap is crucial for equitable health outcomes, as current databases are skewed towards Northern-European populations. In Pakistan, pediatric cancer biobanks are non-existent. Indus Hospital & Health Network (IHHN) in Karachi, with its large pediatric cancer unit, aims to establish a biobank to address region-specific pediatric cancer research needs. This manuscript describes the biobank implementation process using implementation science frameworks.

Methods: The pediatric cancer biobank at IHHN collects FFPE specimens for solid tumors, and isolated mononuclear cells from peripheral blood and bone marrow of suspected acute leukemia. Implementation planning workgroups included clinicians, EMR, IT, management, senior leadership, IRB, and external support from UNC and St. Jude Children's Cancer Hospital. The selection of applicable ERIC (Expert Recommendations for Implementing Change) strategies through stakeholder workgroups considered scope, budget, and feasibility, and context. Standard protocols from ISBER and BCNet guided alignment with best practices. IHHN's past experiences and tacit knowledge gained through rapid, successful implementation also facilitated strategy selection. The EPIS framework (exploration, preparation, implementation, sustainment) was used to map and organize the selected intervention strategies.

Results: Biobank implementation at IHHN, organized by EPIS stages, has been described through a set of 41 implementation strategies. Of these, 34 were selected out of 73 originally published ERIC strategies, while 7 were added based on contextually based workgroup consensus. 599 acute leukemia and 1137 solid tumor specimens have been banked since inception of the biobank operations 2 years earlier. The implementation activities and challenges described include infrastructure, swift specimen collection, prior to treatment, and informed consent. The ancillary processes including training and quality control have also been described and related data presented.

Conclusion: The implementation of Pakistan's first acute leukemia biobank using ERIC and EPIS frameworks offers a structured approach beneficial for settings with limited biobanking experience. This intervention aligns with recognized implementation science frameworks, while addressing aspects pertinent in low- and middle-income countries.

Fig 1. Patient-tracking, consent-taking, sample reception, and bone marrow acquisition steps involved in acute leukemia biobanking at IHHN, Karachi.

Fig 2. Distribution of biobank PBMC (a) FFPE (b) and specimens by primary diagnosis.

Fig 3. ERIC strategies mapped to EPIS process framework, as applied to the pediatric cancer biobank at Indus Hospital & Health Network (IHHN).

Strategies not listed as such in the original work but found useful in our implementation experience have been highlighted.

Fig 4. Quality metrics of PBMC specimens at IHHN, Karachi.

Collection to storage intervals (a) are less than 24 hours for most specimens at present, which is an optimal interval reported by many studies for most functions, though the ideal time reported by others is up to 8 hours following collection. Currently, specimens received before 5 pm are being stored within 8 hours of collection, while this limit is exceeded with specimens arriving after 5 pm as these need to be processed the next day. Collection-storage intervals exceeding 24 hours have mostly been for initial specimens when sample flow, and other protocols were less well defined. Trypan blue viability prior to storage (b) is above 90% for most specimens, but a range of pre-storage viabilities are nevertheless observed. All specimens are stored as these may have different potential uses. In our experience, cells with up to 70% pre-storage viability give optimum RNA yield and quality for transcriptome sequencing, while it is possible that for more sensitive uses like cell culture, specimens with higher viability will need to be selected. Spectrophotometrically determined wavelength ratios (c) for extracted RNA from specimens shows 260/280 ratio of 1.8-2, and 260/230 of more than 2 for most specimens.