Addressing sub-optimal clinical trial design in pancreatic cancer

28th November 2017 (Last Updated November 28th, 2017 17:08)

Progress in pancreatic adenocarcinoma trials is sluggish, which is often attributed to low clinical trial recruitment and substandard design.

Enrolling patients in clinical trials is the preferred treatment channel for all stages of pancreatic adenocarcinoma over standard or accepted therapy, according to the National Comprehensive Cancer Network (NCCN). However, progress in pancreatic adenocarcinoma trials is sluggish, which is often attributed to low clinical trial recruitment and substandard design. This results in an inadequate understanding of this notoriously difficult-to-treat disease and poor outcomes for patients.

Clinical trial accrual is poor in pancreatic cancer and there is a need to enroll patients faster to establish the benefit of experimental drugs within a shorter timeframe. In 2011, 4.6% of the total patient population with pancreatic cancer was enrolled in US-based clinical trials. Set against the clinical trial accrual rate in oncology, which constitutes less than 3% of adult US patients, this rate indicates that comparatively more pancreatic cancer patients enroll in a trial than other cancers. However, the number of patients required to complete accrual in ongoing pancreatic cancer far superseded the number of patients actually enrolling; for instance, a mere 15% of the target enrollment rate was achieved in trials opened in 2011, according to a recent meta-analysis.

Moreover, there was heterogeneity in the supply and demand between different disease stages in pancreatic cancer. From an estimation of available patients on a yearly basis, adjuvant and neoadjuvant trials would require accruing 4.2 times more patients to completely fulfill the needs of enrolling trials, in comparison to 1.3 times more patients in the metastatic setting. Patient barriers to trial entry include inability to meet eligibility criteria and inadequate insurance coverage, especially in markets like the US where a considerable proportion of patients is covered by private insurance plans. In some cases, patients are treated immediately after diagnosis, without comprehensive consideration of all available options, making these patients ineligible for trials in the treatment-naïve setting. The slow accrual rate is a major cause of trials missing their planned completion dates, and even more unfavorably, trial termination due to insufficient number of patients enrolled. Programs providing the appropriate resources and guidelines are required to support accrual to clinical trials; for instance, the National Cancer Institute (NCI) implemented the AccrualNet program in 2010 in the US and in the UK, the National Institute for Health Research (NIHR) Cancer Research Network initiated a program in 2011. Both the US and UK programs successfully increased patient recruitment in clinical trials.

Poor success rates

In addition to poor recruitment, pancreatic cancer trials have notoriously poor success rates. Between 1997 and 2015, from a total of 35 unique therapeutic regimens evaluated in 39 Phase III pancreatic adenocarcinoma trials, only 11% had favorable outcomes. Notably, 85% of these Phase III trials were conducted despite preceding Phase II studies not meeting their primary endpoints. As survival is measured in months rather than years, early determination of whether a treatment is succeeding or failing is required. These shortcomings translate into difficulties in conducting large-scale pivotal trials.

Issues related to clinical trial success can be addressed by both accurate prognosis and careful patient monitoring of treatment efficacy. For example, assessment of circulating tumor DNA or by imaging modalities, such as 18F-fluorodeoxyglucose-PET for tumor metabolism or 18F-fluorothymidine-PET for DNA synthesis, are currently being investigated as techniques that determine early responses. At the moment, it is still questionable whether these approaches will be incorporated into future clinical trial design due to technological limitations. The evaluation of response to treatment within a shorter time interval can provide confidence in performing a clinical trial, particularly if an agent is explored in combination with a standard-of-care option. Moreover, it will also safeguard that a therapy is explored only in the subset of patients proven to have a benefit. Biovica’s DiviTum, an assay determining thymidine kinase activity in a blood sample, was prognostic for overall survival in both metastatic and preoperative patients. This dynamic biomarker proved to be an expedient, non-invasive instrument for predicting outcome and monitoring efficacy in solid tumors.

Early determinants of response and prognostic measures are still lacking in pancreatic cancer. The downturn in clinical trials can be addressed by embracing technological advances, by tackling inadequate operational procedures, and improving patient-physician communication. Lastly, successful strategies in support of accrual to clinical trials require integrating communications between federal agencies for cancer research, such as NCI and NIHR, healthcare agencies, the pharmaceutical industry, and patient advocacy groups.