← MedLegal AI · Deposition Question Guides · Examination Builder · Blog

Questions to Ask an Oncology Expert Witness at Deposition (2026 Guide)

By John Mahoney · Updated 2026-09-01 · 18 questions in 4 concession chains

You are about to cross-examine someone with two decades of training you don't have, on their home turf, about the thing they do every day. That asymmetry is the defining problem of medical-expert depositions — and it gets worse in multi-defendant malpractice cases, where two to four different specialties can be in play in a single case and you have to be conversant in all of them by deposition day.

Oncology experts — usually hematologist-oncologists — testify on delayed cancer diagnosis, staging, treatment toxicity, and febrile neutropenia. Causation is where these depositions are won and lost, and the discipline is to lock qualitative concessions rather than duel over survival statistics.

The answer to the asymmetry is not to learn oncology — it is structure. A concession funnel asks the expert to agree, one fact at a time, to propositions so orthodox that refusing them costs more credibility than granting them. The questions below are a generic-case-safe starting bank, organized into the four chains that structure works through.

How to use these questions

Every question below follows three rules of expert cross-examination: it is leading (it states a fact and asks for agreement), it contains one fact (no compounds — a compound question lets the witness answer the easy half), and it belongs to a concession chain (each answer narrows the room the next answer can live in). Ask them in order. When you get the concession, stop — do not ask the "why" question that lets the expert explain it away.

Chain 1 — Qualifications, retention & scope

Goal: establish on the record that this is a paid, retained witness who never touched the patient, and fix the boundaries of what they actually practice.

  1. Doctor, you were retained by opposing counsel in this case, correct?
  2. You are being compensated for your time reviewing this case and for your testimony today, correct?
  3. You never examined or treated the patient in this case, correct?
  4. Everything you know about this case comes from the materials counsel selected and sent to you, correct?
  5. Oncology treatment decisions are commonly made through multidisciplinary review — surgeons, radiation oncologists, and medical oncologists together, correct?

Chain 2 — Bases-of-opinion lockdown

Goal: freeze the universe. After this chain, any new document, new basis, or new opinion at trial is impeachment material.

  1. You have now told us every document and material you reviewed in forming your opinions, correct?
  2. You are not relying on anything outside that file for the opinions you will give in this case, correct?
  3. You have told us every opinion you intend to offer at trial, correct?
  4. You agree that an opinion you cannot state to a reasonable degree of medical probability is speculation, correct?

Chain 3 — Substantive concessions: earlier is better, staging, and toxicity

No credible oncology expert can reject the propositions below. They are deliberately qualitative — the numbers duel comes at trial, from the literature, not from a deposition shouting match.

  1. In general, cancer is more treatable when it is caught earlier, correct?
  2. Stage at diagnosis drives both the treatment options and the prognosis, correct?
  3. Serial imaging is designed to show change over time, including tumor growth, correct?
  4. Chemotherapy agents have recognized, serious toxicities, correct?
  5. That is why chemotherapy dosing is calculated and verified rather than estimated, correct?
  6. Fever in a chemotherapy patient with a suppressed immune system is treated as an emergency, correct?

Chain 4 — Closing the doors

Goal: end with the record sealed. These questions convert silence into a commitment that nothing new is coming at trial.

  1. You are not offering any opinions in this case beyond the ones we have discussed today, correct?
  2. You have not held back any opinion or criticism that you intend to offer at trial, correct?
  3. You are not offering any opinion about the radiology interpretations in this case, correct?

These are the generic questions. Your case isn't generic.

The MedLegal AI Examination Builder generates a complete, case-specific concession-funnel roadmap for deposing an oncology expert — built from your theory of the case, aware of which side you're on and what kind of witness you're facing, with stop markers, if-they-resist branches, and the doors you should not open. Three free cases, no credit card.

Build my examination → Oncology depo trainer

Where lawyers get burned deposing oncology experts

Dueling over survival percentages

Defense oncology experts live in the survival tables, and a deposition is the worst place to fight them there. Take the qualitative concessions — earlier is better, stage drives prognosis — and save the specific literature for your own expert and for cross at trial, where you control the exhibits.

Letting the expert speculate about growth rates

Experts back-calculate tumor timelines to move the cancer’s staging into a period that defeats causation. Pin the basis: is the growth-rate assumption from this patient’s own serial imaging, or from generalized assumptions? An opinion resting on assumed growth kinetics is built for a Daubert motion.

Ignoring the loss-of-chance framework until too late

Causation standards in delayed-diagnosis cancer cases vary by jurisdiction. If you let the expert opine under the wrong framework unchallenged, the transcript may be useless on summary judgment. Know your state’s standard before the deposition and make the expert commit under it.

FAQ

What questions should I ask an oncology expert witness at deposition?

Work in four concession chains: (1) qualifications, retention, and scope — establish the expert was retained, is compensated, never treated the patient, and what they do and do not practice; (2) bases-of-opinion lockdown — pin every document reviewed and every opinion they intend to offer; (3) the oncology-specific substantive chain — well-established propositions such as "In general, cancer is more treatable when it is caught earlier"; and (4) closing the doors — confirm on the record that no undisclosed opinions remain. Every question should be leading, contain one fact, and build on the prior answer.

How do I cross-examine a medical expert when I never went to medical school?

You don't out-doctor the doctor — you build concession chains from propositions so well-established that a credible oncology expert cannot refuse them, then apply those concessions to the facts in the chart. The expert's medical training becomes your asset: the more orthodox the proposition, the more it costs them to fight it. The asymmetry problem is real — especially in multi-defendant cases with two to four specialties — which is why the questions are organized as funnels rather than as debates.

What are the biggest traps when deposing an oncology expert witness?

The classic traps are dueling over survival statistics at deposition instead of banking qualitative concessions, letting the expert back-calculate tumor growth from generalized assumptions rather than this patient’s own imaging, and letting causation opinions be given under the wrong loss-of-chance framework for your jurisdiction.

Related resources

This page is informational only and is not legal advice. Question banks are generic-case-safe starting points — adapt every question to the facts, the jurisdiction, and your theory of the case before using it in a deposition.