Cancer clinical trials are the engine of oncology progress — every treatment currently used to fight cancer was first tested in a clinical trial before it became standard care. Yet fewer than 1 in 10 eligible adult cancer patients in the United States ever joins a trial, most often because they were never offered one, did not know one existed, or misunderstood what participation involves.
Clinical trials are not a last resort. They are a legitimate, often advantageous treatment option — and in many cases, the only way to access a drug that may work better for your specific tumor’s molecular profile than anything currently approved. Every trial participant has full legal rights, the experimental drug is always provided at no cost, and the Affordable Care Act mandates that insurance cover routine care costs. This article explains the four phases, how to find a matching trial, what enrollment involves, and what the most promising new approaches being tested now look like. For an overview of established cancer treatments, see the cancer treatment guide.
The 4 Phases of Cancer Clinical Trials
Every drug that reaches cancer patients today passed through four development stages. Understanding these phases helps you evaluate the risk-benefit profile of any trial you might consider.
Safety & Dose Finding
20–80 patients; first-in-human; establishes safe dose (MTD/RP2D); highest uncertainty but biomarker-selected Phase I can have striking results
Efficacy Signal
50–300 patients; ORR or PFS primary endpoint; single-arm or randomized; supports FDA Accelerated Approval on surrogate endpoint
vs. Standard of Care
Hundreds to thousands; randomized; OS or PFS primary endpoint; control arm = current best standard treatment; generates regulatory approval data
Post-Marketing
After FDA approval; long-term safety; rare adverse events; new populations; confirms Accelerated Approval surrogate endpoint with OS data
Phase I — Not Always a Last Resort
Phase I trials establish the safe dose. Historically they enrolled patients for whom all standard treatments had failed — making them a genuine last resort. That has changed. Modern Phase I trials testing targeted therapies in biomarker-selected populations can have striking response rates. Selpercatinib (a RET kinase inhibitor) achieved a 68% objective response rate in RET fusion-positive NSCLC in the LIBRETTO-001 Phase I/II trial (NEJM 2020). Larotrectinib, the first NTRK inhibitor, achieved 75% ORR across 12 tumor types in its Phase I/II trial (Drilon A et al., NEJM 2018) — leading directly to FDA tissue-agnostic approval.
These are not typical Phase I numbers. But they illustrate that when your tumor carries a specific molecular alteration and a Phase I trial targets exactly that alteration, early-phase participation is not desperation — it may be the most rational option available. Phase I trials carry the greatest uncertainty (unknown full safety profile; efficacy not the primary endpoint), and patients considering them should discuss the specific mechanism of action, available preclinical and animal data, and whether their tumor harbors the intended molecular target.
Phase II — The Efficacy Signal That Drives Accelerated Approval
Phase II asks: does this drug work in the intended population? The primary endpoint is usually ORR or PFS, measured in 50 to 300 patients selected by tumor type and biomarker. Phase II data supports the FDA’s Accelerated Approval pathway — conditional approval based on a surrogate endpoint (ORR or PFS) with a post-approval requirement for confirmatory Phase III OS data.
Pembrolizumab received its first tissue-agnostic FDA approval in May 2017 based on Phase II data from KEYNOTE-158 — an ORR of 39.6% in MSI-H/dMMR tumors across 10 tumor types, the first tumor-agnostic cancer drug approval in history. For patients, Phase II enrollment means the treatment has passed Phase I safety screening and there is preliminary evidence it works in tumors like yours.
Phase III — The Phase With the Most Favorable Patient Risk Profile
Phase III trials are the pivotal randomized studies that generate definitive FDA approval. Hundreds to thousands of patients are randomized between the experimental treatment and the current standard of care, with overall survival (OS) or PFS as the primary endpoint.
Phase III trials that changed oncology practice:
| Trial | Comparison | Result | Impact |
|---|---|---|---|
| PACIFIC | Durvalumab vs. placebo post-CRT in stage III NSCLC | PFS HR 0.52; OS HR 0.68 | Standard of care for unresectable stage III NSCLC globally |
| KEYNOTE-189 | Pembrolizumab + chemo vs. chemo alone in 1L nonsquamous NSCLC | OS HR 0.56 (updated analysis) | Standard of care 1L nonsquamous NSCLC worldwide |
| DESTINY-Breast04 | T-DXd vs. chemotherapy in HER2-low metastatic breast cancer | PFS HR 0.50; OS HR 0.64 | Created the HER2-low treatment category; opened eligibility to ~50% of HR+ mBC patients |
| SOLO-1 | Olaparib maintenance vs. placebo in BRCA-mutated advanced ovarian cancer | PFS HR 0.30; 3-yr PFS 60% vs. 27% | Standard maintenance therapy for BRCA1/2-mutated advanced ovarian cancer after 1L platinum |
Phase III enrollment is generally the most favorable phase for patients: the drug has passed Phase I safety and Phase II efficacy screening, and you are guaranteed a treatment of established efficacy on either arm.
Types of Trials: Basket, Umbrella, and Platform Designs
Beyond the four phases, modern cancer trials differ in their structural design — and these newer designs are increasingly important for patients whose tumors carry actionable molecular alterations.
Basket Trials (Tumor-Agnostic)
A basket trial tests one drug in patients who share a molecular alteration regardless of where their cancer originated. The cancer type is irrelevant — the molecular target is the selection criterion. KEYNOTE-158 enrolled patients with MSI-H/dMMR tumors across 10 different cancer types — colorectal, endometrial, gastric, biliary, pancreatic, NSCLC, and others — all treated with the same drug. The May 2017 FDA approval of pembrolizumab for MSI-H/dMMR any tumor type was the first ever tumor-agnostic cancer drug approval.
Basket trials are especially valuable for patients with rare tumors that would never generate enough cases for a single-histology trial. If your tumor carries an NTRK fusion, RET fusion, BRAF V600E, or is TMB-H, there may be an active basket trial for which you qualify regardless of organ of origin. Molecular profiling is the prerequisite — see the precision medicine cancer article for how comprehensive genomic profiling identifies these alterations.
Umbrella and Platform Trials
Umbrella trials focus on one cancer type but assign patients to different treatment arms based on molecular subtype. LUNG-MAP is the largest oncology master protocol — a perpetually enrolling umbrella trial for patients with squamous cell NSCLC who have progressed on first-line therapy, with multiple sub-studies matching patients to targeted agents based on their tumor’s actionable alterations.
Platform trials — such as I-SPY 2 (breast cancer neoadjuvant) and STAMPEDE (prostate cancer) — maintain perpetual enrollment, add and drop experimental arms based on interim results, and use Bayesian adaptive randomization to shift enrollment toward more effective arms. Their shared control arm makes them more efficient than running multiple independent trials. A new drug can join an existing platform without starting from zero.
How to Find a Cancer Clinical Trial
Finding a trial that matches your cancer type, molecular profile, prior treatments, and location requires systematic searching through multiple channels.
- Get comprehensive molecular profiling. Many trials require documentation of a specific alteration: EGFR exon subtype, KRAS G12C, HER2 amplification level, MSI-H/dMMR, TMB score, NTRK/RET fusion, PD-L1 TPS, BRCA somatic or germline. Next-generation sequencing or CGP of your tumor tissue — or liquid biopsy at progression — identifies all actionable alterations and opens trial eligibility you might not otherwise know about.
- Search ClinicalTrials.gov. Filter by condition (cancer type), status (recruiting), location (distance from ZIP code), phase, and age. Use Advanced Search to add biomarker or molecular eligibility terms. Read the inclusion and exclusion criteria carefully — these determine whether the trial can consider you.
- Use the NCI trial search and patient navigation line. The NCI clinical trials search provides a simplified interface focused on NCI-funded and NCI-designated cancer center trials. The NCI Cancer Information Service (1-800-422-6237) provides free trial navigation assistance — trained specialists who search for matching trials and help initiate contact with trial sites.
- Contact disease-specific patient advocacy organizations. Leukemia & Lymphoma Society (LLS), Pancreatic Cancer Action Network (PanCAN), and Lung Cancer Research Foundation (LCRF) each maintain trial matching services and staff with disease-specific trial expertise.
- Ask your oncologist for a tumor board referral. Oncologists at NCI-designated comprehensive cancer centers have access to institutional and consortium protocols not yet publicly listed. A multidisciplinary molecular tumor board can identify trial eligibility based on your full genomic profile and make specific trial referrals. For more on tumor boards, see the personalized cancer treatment article.
Understanding Eligibility Criteria
Inclusion criteria typically require: specific tumor type and histology; documented molecular biomarker (the test result must be on file); ECOG performance status (usually PS 0–1 or 0–2); adequate organ function (kidney, liver, bone marrow, cardiac); limited number of prior treatment lines (some trials are 1st line, others 2nd line or beyond).
Exclusion criteria commonly include: prior treatment with the same drug class (prior anti-PD-1 may exclude from IO trials); active or untreated brain metastases; active autoimmune disease (immunotherapy trials); pregnancy; concurrent active malignancy; prior severe hypersensitivity to a similar agent.
Not meeting eligibility criteria for one trial does not mean no trial exists — it means that specific trial was not designed for your exact situation. A different phase, a different molecular arm, or a different drug targeting the same alteration may have different criteria. For questions to bring to any trial evaluation conversation, see the cancer diagnosis questions guide.
What to Expect When You Join a Trial
The Enrollment Process, Step by Step
- Pre-screening. Your oncologist or a trial navigator reviews eligibility criteria against your records before committing to a formal screening visit — avoiding unnecessary travel or testing if obvious disqualifiers exist.
- Referral to the trial site. You submit medical records, pathology slides (sometimes centrally re-reviewed), imaging, and molecular reports to the research coordinator at the trial site.
- Screening visit. Protocol-required tests confirm eligibility: additional labs, sometimes a new biopsy, imaging, EKG, pulmonary function. All paid by the trial sponsor.
- Informed consent. You review and sign the consent document with a research team member before any protocol procedures. Take the document home; bring a family member; ask every question. You keep a signed copy.
- Treatment and protocol visits. Treatment per protocol schedule; labs and safety assessment every 2–4 weeks; response imaging every 6–8 weeks. Report any new symptoms or medications to the research team immediately — not at the next scheduled visit.
- Off-study. You may come off the trial at progression, toxicity, your own decision to withdraw, or a sponsor decision. Coming off the trial does not affect your access to standard cancer care.
What the Trial Pays vs. What Insurance Covers
| Category | Who Pays | Details |
|---|---|---|
| Investigational drug | Trial sponsor (always free) | The experimental treatment is never charged to you or your insurance — ever |
| Protocol-required tests | Trial sponsor | Screening labs, protocol imaging, research biopsies, PK blood draws, study visits |
| Routine care costs | Insurance (required by ACA § 2709) | Standard imaging, hospitalizations, standard labs ordered as part of your routine oncology care |
| Medicare patients | Medicare (required by 42 CFR 411.15(o)) | Medicare must cover routine costs for beneficiaries in qualifying clinical trials |
Before enrolling, request pre-authorization in writing from your insurer and have the trial’s financial coordinator review the protocol coverage breakdown. Most trial sites have financial counselors experienced with this process.
Your Rights as a Trial Participant
Oversight and Safety Monitoring
Every trial site has an Institutional Review Board (IRB) — an independent ethics committee that reviews and approves the trial protocol, consent forms, and ongoing conduct. A separate Data Safety Monitoring Board (DSMB) of independent statisticians reviews interim safety and efficacy data at prespecified intervals and can recommend early trial termination if a treatment causes unexpected harm — or, if efficacy is so compelling that continuing to withhold it from the control arm would be unethical.
Frontiers: What Is Being Tested Now
Personalized mRNA Cancer Vaccines
The KEYNOTE-942 Phase IIb trial (Patel MR et al., NEJM 2023) tested an individualized mRNA neoantigen vaccine (mRNA-4157/V940 — developed by Moderna and Merck) combined with pembrolizumab versus pembrolizumab alone in patients with resected high-risk Stage III/IV melanoma. At 18-month follow-up, the combination showed a 44% reduction in recurrence or death (RFS HR 0.561) — a signal being confirmed in the Phase III KEYNOTE-1089 trial now enrolling. Unlike traditional vaccines that target fixed shared antigens, individualized neoantigen vaccines are sequenced from each patient’s own tumor, identify unique mutations, and are manufactured as a custom mRNA construct within weeks of tumor resection. This represents the intersection of immunotherapy and precision medicine at its most individualized extreme. For broader immunotherapy context, see the cancer immunotherapy article.
Radioligand Therapy
Lutetium-177 PSMA-617 (Pluvicto) demonstrated significantly improved OS in metastatic castration-resistant prostate cancer in the VISION trial (Sartor O, NEJM 2021), leading to FDA approval in March 2022. Clinical trials now evaluate lutetium PSMA in earlier disease settings (PSMAddition in metastatic hormone-sensitive prostate cancer; PSMAfore pre-docetaxel mCRPC) and in combination with PARP inhibitors and immunotherapy. Additional radioligand targets in early development include DLL3 for SCLC, HER2 for breast cancer, and FAPI for multiple solid tumor types.
Bispecific T-Cell Engagers in Solid Tumors
Tarlatamab, a DLL3-CD3 bispecific T-cell engager, achieved a 40% objective response rate in heavily pre-treated extensive-stage SCLC (DeLLphi-301 Phase II, Ahn MJ et al., NEJM 2023) — a cancer type with very limited options after second-line therapy. FDA granted Breakthrough Therapy designation; Phase III is enrolling. Amivantamab (EGFR×MET bispecific) combined with lazertinib is being evaluated in EGFR-mutated NSCLC after osimertinib resistance.
KRAS-Targeted Combinations
Sotorasib and adagrasib were the first approved KRAS G12C inhibitors, but resistance emerges rapidly through compensatory RAS-MAPK pathway reactivation. Combination strategies now in trials: adagrasib + SOS1 inhibitor (TNO155); adagrasib + cetuximab (KRYSTAL-10 for KRAS G12C CRC); next-generation KRAS G12C inhibitors with improved selectivity. First-in-class KRAS G12D inhibitors (targeting one of the most prevalent KRAS mutations in pancreatic and colorectal cancer) are entering Phase I — potentially addressing one of the last major un-druggable oncogenes.
Frequently Asked Questions
- Drilon A et al. — Larotrectinib in NTRK fusion-positive cancers; NEJM 2018
- Marabelle A et al. — KEYNOTE-158 MSI-H pembrolizumab tissue-agnostic; Lancet Oncol 2020
- Antonia SJ et al. — PACIFIC (durvalumab stage III NSCLC); NEJM 2018
- Gandhi L et al. — KEYNOTE-189 (pembrolizumab + chemo 1L NSCLC); NEJM 2018
- Modi S et al. — DESTINY-Breast04 (T-DXd HER2-low breast cancer); NEJM 2022
- Moore K et al. — SOLO-1 (olaparib BRCA-mutated advanced ovarian cancer); NEJM 2018
- Drilon A et al. — LIBRETTO-001 (selpercatinib RET fusion NSCLC); NEJM 2020
- Ahn MJ et al. — DeLLphi-301 (tarlatamab SCLC bispecific); NEJM 2023
- Sartor O et al. — VISION (lutetium-177 PSMA-617 mCRPC); NEJM 2021
- Patel MR et al. — KEYNOTE-942 / mRNA-4157 individualized neoantigen vaccine; NEJM 2023
- ACA § 2709 — Clinical trial insurance coverage requirement
- Right to Try Act, 2018 — US federal law on investigational drug access
- ClinicalTrials.gov — clinicaltrials.gov
- NCI Clinical Trials Search — cancer.gov/research/participate/clinical-trials-search
This article is for educational purposes only and does not constitute medical advice. Discuss all cancer treatment and clinical trial decisions with your oncology care team.