Shrinking oncology phase I clinical trials: If the drug is a breakthrough, why would you ever want to utilize the maximally tolerated dose?

Share on facebook
Share on twitter
Share on linkedin
Share on email
Share on print

Historically, oncology drug development has evolved on what may seem to be a different planet, at least relative to mainstream clinical pharmacology.

Whereas traditional drug development utilizes standard tools of randomization, pharmacokinetic data, and blood-based biomarkers to optimize dosing, oncology has primarily utilized safety data (i.e., maximally tolerated dose) to guide dose selection.

In this context, I and others have advocated for the use of expansion cohorts to try to precisely define the maximally tolerated dose, although such estimates are often unreliable due to small cohort sizes (1-4).

While this historical paradigm may have been appropriate for cytotoxic chemotherapy agents, it is certainly not appropriate for precisely targeted enzyme inhibitors or monoclonal antibodies, comprising much of modern oncology drug development.

The lack of proper dose selection in oncology has led to a number of post-marketing requirements to study lower doses, with many of these studies leading to label changes (5). It has also created an opportunity to reduce prescribing costs using lower dosages or less frequent dosing off-label (6-10).

The few studies published to date represent only a small fraction of the opportunity to reduce prescribing costs through off-label dosing, an opportunity created by the industry’s lack of rigor in dose selection(11), and leading to the formation of the Value in Cancer Care Consortium.

A recently published FDA draft Guidance (12) now seeks to advise companies on approaches to optimize phase I clinical trial designs with particular attention to the increasing and widespread use of expansion cohorts. While concerns regarding the size of some phase I trials have been longstanding (13), the recent draft Guidance may have been motivated by Merck’s phase I study of pembrolizumab, comprised of 1,235 patients (14).

The draft Guidance augments the information in the Code of Federal Regulations (21CFR312.21) that explicitly articulates the aims and appropriate sizes of phase I, II, and III trials. Notably, phase I studies are stated to be “generally in the range of 20 to 80”, given the goal of obtaining “sufficient information about the drug’s pharmacokinetics and pharmacological effects… to permit the design of well-controlled, scientifically valid, phase II studies.” In this context, there is no suggestion that phase I studies can or should identify the optimal dose, and in fact this is generally (i.e., outside oncology) ascertained in the context of randomized dose-ranging phase IIb studies (15).

The draft Guidance appears to suggest that expansion cohorts are overused, and should only be considered in the context of a breakthrough drug. It logically follows that in the absence of clinical activity suggesting breakthrough activity, that the sponsor should follow a more traditional approach of completing and fully analyzing a phase I trial before beginning phase II studies.

The draft Guidance also articulates reasons not to utilize expansion cohorts even if the drug demonstrates potential breakthrough activity, such as steep toxicity indices or high interindividual or intraindividual variability. Given the need to exclude high variability, this suggests that expansion cohorts should not be initiated until there are sufficient pharmacokinetic data available, including data regarding intraindividual variability, a particular problem for many small molecule kinase inhibitors (16).

Furthermore, the draft Guidance explains that should a sponsor wish to conduct one or more expansion cohorts, that each of these have a clear scientific rationale and statistical plan, and that FDA input should be sought early and often. Notably, the draft Guidance also raises concerns regarding the adequacy of safety monitoring for expansion cohorts, and mandates the incorporation for an independent safety assessment or data monitoring committee for all first-in-human multiple expansion cohort protocols.

Thus, my interpretation of the FDA’s position is that protocols with expansion cohorts should be the exception and not the rule, that FDA review of such protocols is required, and that if there truly is a compelling reason to study multiple expansion cohorts, that there will be many additional study requirements in the interest of both protecting patients and generating scientific knowledge.

If my interpretation is correct, I am enthusiastically supportive of the FDA’s recommendations. Phase I trials have gone from being proper clinical pharmacology studies conducted by experienced investigators at single institutions to massive multi-center studies run by contract research organizations, and often overseen by medical monitors with limited experience in drug development. The site investigator’s responsibilities are limited to enrolling patients and completing case report forms. Manuscripts are drafted by a scientific writer employed or contracted by the sponsor, and peer reviewers are generally more focused on efficacy results than the clinical pharmacology of the agent under study.

My hope is that the draft Guidance will be finalized soon with minimal or no revision, and that sponsors will be forced to revisit the historical principles of clinical pharmacology that are widely utilized in all therapeutic areas other than oncology. And to the sponsors out there: if your drug is so good, why would you ever want to utilize the maximally tolerated dose?


  1. Mick R, Ratain MJ. Model-guided determination of maximum tolerated dose in phase I clinical trials: evidence for increased precision. J Natl Cancer Inst. 1993;85(3):217-23. PubMed PMID: 8423626.

  2. Hamberg P, Ratain MJ, Lesaffre E, Verweij J. Dose-escalation models for combination phase I trials in oncology. Eur J Cancer. 2010;46(16):2870-8. doi: 10.1016/j.ejca.2010.07.002. PubMed PMID: 20691584.

  3. Iasonos A, O’Quigley J. Design considerations for dose-expansion cohorts in phase I trials. J Clin Oncol. 2013;31(31):4014-21. doi: 10.1200/JCO.2012.47.9949. PubMed PMID: 24101039; PMCID: PMC4979131.

  4. Sharma MR, Ratain MJ. Taking a Measured Approach to Toxicity Data in Phase I Oncology Clinical Trials. Clin Cancer Res. 2016;22(3):527-9. doi: 10.1158/1078-0432.CCR-15-2005. PubMed PMID: 26467385; PMCID: PMC4738103.

  5. Lu D, Lu T, Stroh M, Graham RA, Agarwal P, Musib L, Li CC, Lum BL, Joshi A. A survey of new oncology drug approvals in the USA from 2010 to 2015: a focus on optimal dose and related postmarketing activities. Cancer Chemother Pharmacol. 2016;77(3):459-76. doi: 10.1007/s00280-015-2931-4. PubMed PMID: 26811176; PMCID: PMC4767861.

  6. Lichter AS. From Choosing Wisely to Using Wisely: Increasing the Value of Cancer Care Through Clinical Research. J Clin Oncol. 2018;36(14):1387-8. doi: 10.1200/JCO.2018.78.4264. PubMed PMID: 29590009.

  7. Ratain MJ, Chen LS, Lichter AS, Harvey RD, Smith SM, Saltz LB, Wierda WG, Gandhi VV, J. KM. Sales revenues at the potential expense of patient safety: the example of You&iTM. The Cancer Letter. 2018.

  8. Ratain MJ, Goldstein DA. Time Is Money: Optimizing the Scheduling of Nivolumab. J Clin Oncol. 2018:JCO1800045. doi: 10.1200/JCO.18.00045. PubMed PMID: 30148658.

  9. Schiffer CA. The evolution of dasatinib dosage over the years and its relevance to other anticancer medications. Cancer. 2018;124(13):2687-9. doi: 10.1002/cncr.31516. PubMed PMID: 29723416.

  10. Szmulewitz RZ, Peer CJ, Ibraheem A, Martinez E, Kozloff MF, Carthon B, Harvey RD, Fishkin P, Yong WP, Chiong E, Nabhan C, Karrison T, Figg WD, Stadler WM, Ratain MJ. Prospective International Randomized Phase II Study of Low-Dose Abiraterone With Food Versus Standard Dose Abiraterone In Castration-Resistant Prostate Cancer. J Clin Oncol. 2018;36(14):1389-95. doi: 10.1200/JCO.2017.76.4381. PubMed PMID: 29590007; PMCID: PMC5941614.

  11. Ratain MJ, Lichter AS. Potential for value-based prescribing of oral oncology drugs. ESMO. 2018.

  12. FDA. Expansion Cohorts: Use in first-in-human clinical trials to expedite development of oncology drugs and biologics guidance for industry DRAFT GUIDANCE 2018. Available from:

  13. Tolcher AW, Takimoto CH, Rowinsky EK. The multifunctional, multi-institutional, and sometimes even global phase I study: a better life for phase I evaluations or just “living large”? J Clin Oncol. 2002;20(21):4276-8. doi: 10.1200/JCO.2002.20.21.4276. PubMed PMID: 12409324.

  14. Kang SP, Gergich K, Lubiniecki GM, de Alwis DP, Chen C, Tice MAB, Rubin EH. Pembrolizumab KEYNOTE-001: an adaptive study leading to accelerated approval for two indications and a companion diagnostic. Ann Oncol. 2017;28(6):1388-98. doi: 10.1093/annonc/mdx076. PubMed PMID: 30052728; PMCID: PMC5452070.

  15. Ratain MJ. Targeted therapies: redefining the primary objective of phase I oncology trials. Nat Rev Clin Oncol. 2014;11(9):503-4. doi: 10.1038/nrclinonc.2014.135. PubMed PMID: 25091610.

  16. Chatelut E, Bruno R, Ratain MJ. Intraindividual Pharmacokinetic Variability: Focus on Small-Molecule Kinase Inhibitors. Clin Pharmacol Ther. 2018;103(6):956-8. doi: 10.1002/cpt.937. PubMed PMID: 29194586.


President Joe Biden’s proposed Advanced Research Projects Agency-Health would be a welcome partner to NCI—particularly in conducting large, collaborative clinical investigations, NCI Director Ned Sharpless said.“I think having ARPA-H as part of the NIH is good for the NCI,” Sharpless said April 11 in his remarks at the annual meeting of the American Association for Cancer Research. “How this would fit with the ongoing efforts in cancer at the NCI is still something to work out.”
Mark Ratain
Director of the Center for Personalized Therapeutics, the Leon O. Jacobson Professor of Medicine, and associate director for clinical sciences at the University of Chicago Comprehensive Cancer Center