Teleoncology Implementation in a Low-Resource Setting: A Retrospective Cohort Study From Lubango, Angola
Paulo Salamanca, Maria Lina Antunes, Cristina Cangoma, Teresa Garcia, Irina Bernardo Jacinto, Celeste Muteca, Arminda Sambundo, Madalena Tyvela, Milton Munene, Alcina Campos, Lisandra Rios, Eliane Muachissengue, Ana Rodrigues, Leonardo Barreira, Pedro Cruz, António André, Telma Costa, Ivo Julião, Lúcio Lara SantosBackground:
We evaluated whether structured teleoncology could support the safe and affordable delivery of systemic oncology treatment at a provincial Angolan hospital without such installed capacity previously.
Methods:
Retrospective cohort study at Lubango Central Hospital (HCL), Angola, as a proof of principle, September 2024–May 2025. The initial analysis covered 85 patients discussed at a multidisciplinary tumor board (MTB) by videoconference; a second analysis assessed safety, outcomes, and costs in 23 patients treated at HCL. Efficacy was judged against palliative-intent indicators.
Results:
In a cohort presenting overwhelmingly with advanced disease (87.0% stage III–IV), 93 cycles were delivered under remote supervision with no preparation errors, no contamination, no grade 4–5 adverse events (95% confidence interval 0–4.0%), no toxicity admissions, and no treatment-related deaths. Course completion reached 89.5%. The MTB processed 156 consultations and met all technical and operational targets; grade 1–3 events occurred in 16.1% of cycles, and partial remission reached 52.6%. Symptom onset to first presentation had a median of 6 months (range 1–11.7); the program-attributable interval was 1–2 weeks. Unit cost was 223,000 US dollars (USD); the drug cost was 251.93 USD per patient, against 643–794 USD averted per course in travel, lodging, and subsistence, and 533 USD per cycle in personnel costs (base salary).
Conclusions:
This program showed that systemic treatment under remote supervision was delivered safely and feasibly without a resident medical oncologist, with household savings of two and a half to three times the drug cost. This proof of principle suggests that the model may be workable elsewhere, though the small sample and absence of a control group limit firmer conclusions.