Self-Directed Cancer Treatment - Patient Agency When Standard Care Runs Out

Summary

Sid Sijbrandij (GitLab co-founder) documents his self-directed approach to treating osteosarcoma after exhausting standard care options and finding no available trials. He pursued maximum diagnostics, created new treatments, ran treatments in parallel, and published all his medical data (25 TB) publicly to help others.

Key Insight

• When standard of care runs out and no clinical trials are available, patients can take agency by combining maximum diagnostics with parallel experimental treatments rather than sequential ones
• Sijbrandij published 25 TB of medical data in publicly readable Google Cloud buckets with a full treatment timeline - this radical transparency model could accelerate research for rare cancers where data is scarce
• The approach challenges the typical sequential, institution-controlled treatment model: instead of waiting for one treatment to fail before trying the next, he ran multiple interventions simultaneously
• Ruxandra’s linked article highlights how medical bureaucracy actively blocks patients from accessing potentially life-saving treatments - the regulatory framework optimizes for liability reduction, not patient outcomes
• His OpenAI Forum presentation connects AI tools (ChatGPT specifically) to his cancer fight, suggesting LLMs can help patients navigate complex medical literature, identify treatment options, and coordinate care across institutions
• The “scaling this for others” framing turns a personal health crisis into an open-source-style project, consistent with his background building GitLab