In this conversation, Sean Khozin and Sunil explore the formative experiences that shaped his worldview and leadership philosophy, what it means to build healthcare infrastructure around the concept of healing, and how AstraZeneca's oncology portfolio expanded from a single asset into one of the most consequential in the world.

They go deep on the science and strategy behind antibody-drug conjugates (ADCs), the frontier of ADC and immunotherapy combinations, and a distinction Sunil considers the true next frontier of the field: the difference between precision medicine and genuinely personalized medicine, where patient values are matched to therapeutic value.

In this episode, we discuss:

1. The "Next HIV": Why Dr. Shariff believes cancer care is reaching a similar turning point in long-term management.
2. The Gap in Data: Why traditional clinical trial reporting often fails to capture the nuances physicians see at the bedside.
3. AI & Triage: How unique e-consult models and AI tools are helping oncologists manage complex side effects more efficiently. The Future of Oncology:
4. Predictions on patient-driven care and the push for greater data accessibility to drive innovation.

As a third-year medical student, David developed idiopathic multicentric Castleman disease, a rare inflammatory disorder that led to multi-organ failure. He was read his last rites five times. Facing a condition with no clear therapeutic roadmap, he began banking his own blood, performing proteomic analyses, and identified an mTOR pathway signal that led to a life-saving repurposed transplant drug.

He has now been in remission for over a decade. But survival was only the beginning.

David went on to found the Castleman Disease Collaborative Network and later co-founded Every Cure, a nonprofit using AI and computational pharmacophenomics to systematically evaluate every approved drug against every known disease. This episode is about moving from serendipity to strategy in medicine — and what it will take to build systems that leave no patient behind.

The conversation examines how incentives, reimbursement, and regulation shape the fate of cancer therapies, sometimes in rather unpredictable ways. It also explores Yosemite’s founding by Reed Jobs, Dan’s longstanding friendship with him, and how trust, time horizon, and shared mission have shaped a hybrid model that combines venture investment with grant-based risk-taking.

This is a discussion about cancer, but it is equally a discussion about architecture: how systems are designed and what they reward.

In this episode, we explore the Parker Institute's unique model of "venture philanthropy." Karen explains how PICI uses shared intellectual property to de-risk early-stage science and spin out venture-ready companies like Dispatch and Orbital Therapeutics. We also discuss the challenges of modernizing a historic non-profit with "no product" to sell , the intersection of AI and biomarkers with companies like ArteraAI , and PICI's massive data project, "Radiohead".

Karen is candid about her leadership philosophy: she is not a "keep the trains running" executive, but a leader built for transformation.

Daniel Arbess has built a career on questioning foundational assumptions across multiple domains: from analyzing Cold War nuclear policy to restructuring post-Soviet industries as the youngest-ever partner at White & Case, to founding Xerion Capital, a hedge fund that delivered over 25% annualized returns through the 2008 financial crisis. Now, he's applying that same lens to healthcare transformation.

Dr. Gulley is one of the world’s leading figures in immuno-oncology. He helped bring immune-based cancer therapies from an era of skepticism into a clinical mainstay, shaping both the science and the clinical frameworks that define the field today.

Their conversation explores the critical frontiers of immunotherapy: the limits of predictive biomarkers, the biological opacity of treatment resistance, and the shortcomings of current trial designs. They discuss adaptive clinical trials, new endpoints, and the integration of AI into clinical research. Emerging tools such as digital pathology, wearable monitoring, and continuous physiologic data capture are highlighted, including recent work detecting CAR-T toxicity before it becomes clinically evident.

We discuss her path from autoimmune skin disease research at Stanford to building Medable into a global clinical trial infrastructure platform operating across 14 therapeutic areas, and what it takes to align technology, AI, and scientific rigor with the needs of patients, regulators, and researchers.

In the first episode of Precision Signals, Dr. Sean Khozin speaks with Dr. Cora Sternberg and Dr. Olivier Elemento—two leaders at the intersection of clinical trials, precision medicine, and computational biology.

Together, they unpack the central paradox of medical AI: why so many tools are technically validated, yet so few are clinically trusted. Topics include:

• Why randomized clinical trials—not just retrospective benchmarks—are essential for AI adoption
• The limitations of current training data and the risks of treating clinical guidelines as ground truth
• The evolving role of real-world data, EHRs, and multimodal inputs in building more adaptive AI systems
• What physicians actually need to trust and use AI in everyday cancer care

This is a candid conversation about evidence, trust, and what it will take to build AI that earns a place in the oncology clinic, grounded in insights from the participants’ recent review published in NEJM AI: https://ai.nejm.org/doi/abs/10.1056/A...