Special issues in oncology clinical trials

BRAF targets in melanoma. Biological mechanisms, resistance, and drug discovery. Cancer drug discovery and development. Volume 82. Ed. Ryan J. Sullivan. Springer (2015)

Phase one trials

Early phase clinical trials in oncology have long been assailed as promoting enrollment in an endeavor that inherently has an unfavorable risk-benefit ratio, because the chance for therapeutic benefit is low yet there are significant risks of toxicity and even death from adverse events related to a novel therapy. Given that the primary objective of a phase I clinical trial is to establish safety, short-term toxicities, and maximum tolerated doses of agents in preparation for a phase II trial which would then study efficacy, one can see why phase I trials have a “bad name.” While the concern for real harm exists around phase I trials involving new cytotoxic chemotherapies, the landscape for novel agents has changed considerably in the era of targeted therapies [57]. Another point worth considering is that participation in active treatment, possibly via a phase I trial, may have potential psychosocial benefits that arise from improvements in quality of life [58] as well as the notion of helping future patients suffering from cancer [59]. Another major criticism of phase I oncology trials is the assumption that there may be glaring omissions in the informed consent process and possibly a low quality of disclosure on the part of researchers, given the low potential for benefit and the vulnerability of a patients with incurable cancers.

Interim analyses and early stopping rules

As previously discussed, in phase III RCTs, ongoing monitoring of trial activities is typically overseen by the DSMB, whose functions are to review trial data, including interim analyses, with the goal of determining whether a trial is considered appropriate for continuation based on early stopping rules that involve definitions of unacceptable toxicities or unacceptable differences in certain pre-specified endpoints (typically overall survival). If the DSMB determines that one group of participants has suffered excess toxicity or that there is a significant survival advantage to one treatment over another, the trial is usually halted and participants are offered a chance to receive the more efficacious therapy. This is, in theory, one way to preserve equipoise throughout the life of an RCT, and to protect participants for unnecessary harm while ensuring that they have access to the maximum amount of benefit.

The ethical concerns pertaining to early stopping rules arise from three major areas: the validity of informed consent in light of interim findings, risks and benefits to current and future patients and the broader implications of the trial under question on the field [60]. In the case of informed consent, one can argue it is an ongoing responsibility of investigators (as overseen by an IRB) to fully disclose interim results to patients, who may then exercise their right to terminate participation in a trial. One solution that has been proposed to this dilemma is that informed consent may explicitly state that preliminary analyses will not be disclosed prior to trial completion, which would address the purely regulatory concern [5252, 61]. However, others would argue that asking a participant to relinquish his or her rights to information jeopardizes informed consent, and in some cases may even approach fraud [62].

With respect to the impact of early study termination on reporting of benefits of an investigational regimen, stopping a trial due to an improvement in a surrogate endpoint such as progression free survival (PFS) without an improvement in overall survival (OS) may lead to incomplete data collection on the adverse events and toxicities associated with therapy. It is conceivable that a new regimen may lead to tumor regression initially, but that toxicity may be dose-limiting or even necessitate that patients discontinue therapy, which may in turn negate any potential overall survival benefit. Stephen Cannistra cites an illustrative example in the Journal of Clinical Oncology in 2004 [63]:

  • A phase III RCT of 12 months of maintenance paclitaxel versus 3 months of paclitaxel was performed in patients with advanced ovarian cancer [64]. The primary co-endpoints were PFS and OS. The trial was halted early after interim analysis demonstrated that there was a statistically significant improvement in PFS of 7 months in the 12 month maintenance group without any evidence of an OS benefit [64, 65]. However, a significantly higher proportion of participants in the 12 month arm compared to the 3 month arm had grade 2 or 3 neuropathy (23 % vs. 15 %), and no quality of life measures were assessed [64]. Thus, it is conceivable that participants on the 12-month arm suffered a higher burden of side effects for several months of progression-free survival that may or may not have translated into an OS benefit (had the trial been followed to completion).

Cannistra proposes the following recommendations as a way to frame the discussion about the role of early stopping rules in oncology RCTs [63].

  1. Early stopping rules should be limited to and based on either: unacceptable toxicity or improvements in overall survival or quality of life.
  2. Informed consents should include clear statements regarding criteria for early trial termination in the interest of the most complete disclosure
  3. Investigators and participants should have access to interim analyses reporting on either toxicity or overall survival.
  4. There should be a greater effort to invite patient representation and advocates to serve on DSMBs.

[contact-form-7 id=»5168″ title=»Контактная форма 1″]


Добавить комментарий

Войти с помощью: 

Ваш e-mail не будет опубликован. Обязательные поля помечены *