Eye On Biotech

Hello, my name is Harry Spencer. I studied medicine at university and have always been fascinated by the intersection between academic research and the development of new therapeutics.

After completing my foundation training, I took some time out of medicine to work at a biotech-focused VC fund and began studying for the CFA. I then moved into an operational consulting role, where I first started coding while helping to build a simple regression model to estimate pricing elasticity. I later returned to medicine, initially training in histopathology before moving into ophthalmology.

After coming back to clinical work, I found myself looking for a personal project that could combine my interest in financial analysis with clinical appraisal of new therapeutics. Starting a blog where I could share my research felt like a natural fit. It gives me a place to publish my work, improve through feedback, and experiment with building and maintaining my own website. I enjoy learning about investing and biotech, and I hope this blog can serve as a space to share my thinking and continue improving over time.

Each analysis follows a fixed structure designed to mirror the workflow of a clinical-stage investment analyst. The goal is to separate biological plausibility, trial quality, and commercial viability into distinct assessments before combining them into a final probability estimate.

1. Disease Biology

A summary of the target disease, its pathophysiology, and the current standard of care. This section establishes the clinical context and frames the unmet need the therapy is attempting to address.

2. Mechanism of Action

An evaluation of the biological target, the pathway rationale, and whether the target is validated. Mechanism-specific risks are identified here — not just class effects, but risks specific to the delivery method, molecule type, or therapeutic approach.

3. Trial Design Critique

A structured review of the trial design, including endpoint selection, comparator appropriateness, power assumptions, and sources of potential bias. Regulatory design decisions (superiority vs non-inferiority, primary vs secondary endpoints) are assessed in the context of both approval likelihood and commercial meaningfulness.

4. Endpoint Analysis & Real-World Application

Available data are reviewed and contextualised against the specific endpoints used. Where possible, effect sizes are compared against prior trials and benchmarked against the threshold likely to drive clinical adoption, not just regulatory approval.

5. Safety Profile

Known adverse events are assessed for clinical significance and adoption risk. Particular attention is given to events that could limit real-world use even if a therapy achieves regulatory approval.

6. Competitive Landscape & Adoption Analysis

The therapy is placed in its commercial context: pipeline competitors, prescribing inertia, payer dynamics, and the magnitude of improvement required to shift clinical practice. These sections are intentionally separated from the clinical analysis because commercial success and regulatory success are distinct outcomes.

Probability Framework

A blended probability estimate is produced by multiplying four conditional probabilities: mechanism plausibility, trial success, regulatory approval given a positive trial, and meaningful commercial adoption. Each component is assigned independently before being combined. The final figure is not a price target or investment recommendation — it is a structured estimate of the probability of clinical and commercial success, intended to make my reasoning explicit and falsifiable.

Clinical Materiality Score

A 1–10 composite score across six dimensions: biological plausibility, endpoint strength, effect size relevance, safety profile, adoption friction, and commercial durability. This score reflects how meaningful the therapy would be to patients and clinicians if approved — independent of its probability of reaching that point.