Clinical trials—Designing, implementing, and collaborating

1 |. BACKGROUND/NEED

The Institute of Medicine estimates that 80% to 90% of therapies are not supported by high-quality clinical evidence such as that provided by clinical trials.¹ In oncology, fewer than 1 in 20 adult cancer patients enroll in cancer clinical trials.² A recent report in the Journal of Clinical Oncology compared treatment effects seen in observational studies and randomized controlled trials (RCTs) comparing the same treatment regimens and found no correlation. Observational studies were significantly more likely to demonstrate longer survival with the addition of surgery but no such pattern was identified among RCTs and only 40% of matched studies reported the same conclusion.³ This disparity along with the observation that cancer patients who participate in clinical trials have better outcomes than those who do not, highlight the critical need for novel, impactful clinical trials to improve patient care.⁴

The number of clinical trials in surgery pales in comparison to those in other disease specialties. What compounds the dearth of prospective evidence in surgical care is the lack of well-planned, completed surgical trials. Although >300 surgery trials were registered in clinicaltrials.gov between January 2011 and December 2012, a significant number (72.7%) were associated with research waste, defined as non-publication, inadequate reporting or presence of avoidable design limitations.⁵ This reality may account for a recent statistic that among 9961 oncology-related trials registered between 2001–2011, only 10% included a surgical intervention.⁶

Given the identified importance of clinical trials in the care of cancer patients and the central role of surgery in modern multimodality therapy, educating surgical oncology trainees in the development and assessment of clinical trials is essential. Further, surgeons are becoming increasingly involved in or leading treatment sequencing trials that were historically considered “nonsurgical” such as neoadjuvant and window-of-opportunity trials.⁷ These issues have been recognized by the surgical education community and clinical trial development/assessment are included as milestones for the most recent version of the ACGME—Complex General Surgical Oncology trainee and breast surgical oncology fellowship minimum training requirements. Based on these needs, our Department of Surgical Oncology developed a Clinical Trials Research Curriculum to educate our trainees in the process and critical elements of designing and running surgical clinical trials.

2 |. GOALS AND COURSE STRUCTURE

The initial version of the Surgical Oncology Fellows Clinical Trials Course was modeled after a clinical trial course created for junior faculty at our institution. The “MD Anderson Clinical Trial Methods and Design Workshop for Junior Faculty” is an intensive, 2-day course led by senior faculty and highly experienced clinical investigators, with the goal to assist junior faculty in developing novel concepts into actionable clinical protocols by the completion of the workshop. Our departmental faculty recognized the value of the contents of this faculty course as a learning opportunity for fellows. Thus, we designed the initial iteration of a course to address the needs of next generation of surgical investigators by focusing on clinical trial design and implementation. Stakeholders, including teaching faculty, administrative faculty, and trainees, worked together to identify the optimal course structure for trainees.

Important considerations included accommodating the busy schedule of clinical fellows and providing the educational material in shorter blocks spaced between intervening weeks to address comments from each previous session. It was logistically challenging to secure 2 consecutive days of protected time for fellow engagement and therefore, a modified version of the course was considered and implemented in 2016 (Table 1). The current fellows course includes 3-half day sessions over 3 months combining didactic lectures and interactive small groups to assist fellows in the development of high quality, feasible, and innovative clinical research concepts. Based on course feedback, there have been some modifications in the content and structure, but the underlying goals and objectives remain unchanged.

TABLE 1.

Clinical trials curriculum course structure

Day 1	Day 2	Day 3
Didactic goals
Study design considerations	Logistics	Investigator responsibilities
PICO	Budget/contract negotiation	Staffing
Statistics-basic and Bayesian	Patient-reported outcomes	Collaborators & patient advocates
Components of trial protocol	Cooperative groups/industry partners	Site-initiation visit
Traditional and alternative trial designs	Regulatory (IRB, DSMB, etc)	Accrual
Small group goals
Define feasible research question	Study design schema	Group presentations
Identify study population	Inclusion/exclusion criteria	Colleague feedback
Determine appropriate endpoint	Primary/secondary endpoints

The overarching goal of the MD Anderson Surgical Oncology Fellows Clinical Trials Design Course is to cultivate thought-leaders in clinical research. The educational objectives to achieve this goal are:

• To facilitate the development of clinically relevant prospective trials to improve patient treatment options and outcomes.

• To identify potential barriers to clinical trial activation and completion and identify solutions to overcome barriers

• To define Principal Investigator (PI) responsibilities and identify “best practices” for the conduct of cancer clinical trials.

3 |. COURSE CONTENT

The course uses a combination of didactic and small group sessions, designed as an introductory course focusing on practical design concepts with some instruction related to the logistics of running a clinical trial. Fellows are asked to come to the course with a clinical concept that they would like to develop into a prospective trial and throughout the course utilize the course material to mature this concept.

3.1 |. Didactics

3.1.1 |. Study design considerations

A successful clinical trial begins with choosing the right question. In the first series of lectures, fellows are instructed in the PICO method (ie “patient, intervention, comparison, and outcome”), the different phases of clinical trials, basic trial design including components of a protocol, classical frequentist statistical methods including sample size and power calculations, as well as advanced statistical methods including Bayesian analyses.

External validity, or generalizability, is a major focus of the first session of the course. In the design of their ideas, the trainees are encouraged to think about the clinical relevance of their PICO ideas as well as their anticipated outcomes. If the study does not integrate external validity into its design, then the time, resources, and, most importantly, patients enrolled, are lost into a trial whose results are not likely to change clinical practice.

Alternative trial designs such as n-of-1 trials and cluster-randomized trials are reviewed as well as discussions about the number of research opportunities that exist in everyday practice through the learning healthcare system and pragmatic study design. Traditional design classes and statistics classes focus on “efficacy” trials, often with medical oncology trials as the best examples, in which one drug/combination is compared to a control. Pragmatic “effectiveness” trials focus on the correlation between treatments and outcomes in real-world healthcare systems. Trainees are encouraged to identify quality improvement (QI) processes that can be translated into a prospective process in which the goals are delineated a priori and studied either through the traditional IRB or more often now an institutional QI board. Genomically-informed trials such as basket, umbrella and window-of-opportunity trials with translational correlatives including institutional examples of recently completed and current trials are described. At the end of this session, fellows are asked to consider the importance of their scientific question, whether the preclinical and clinical data are sufficiently strong to justify the rationale, and if the study design will allow the question to be answered.

3.1.2 |. Clinical trial logistics

Once a clearly defined (and potentially-answerable) research question has been developed, determining the feasibility and logistics of running a clinical trial are paramount. Lectures dedicated to budgeting and contract negotiation including institutional overhead costs, administrative fees, salary support, patient care costs and ancillary services are reviewed to inform monetary feasibility. In particular, the costs of running prospective trials can be eye-opening for trainees, but our course also focuses on identifying low-cost ideas for their first trials. Funding for investigator-initiated trials can sometimes be supported by institutional funding mechanisms, but the ability to perform larger trials with either novel therapeutics or novel technology often requires industry partnership. Navigating the financial and ethical issues that arise in clinical trials with industry support are reviewed.

In addition, feasibility of trial accrual including evaluation of competing clinical trials, track record of recruitment, and patient population at the lead institution, are all critical to ensure completion. The ability to execute a well-designed clinical trial relies on a supportive team beyond physician investigators. Information regarding the day-to day management of the trial that is maintained by a team of personnel including but not limited to research nurses, data coordinators and contract specialists, are critical to contribute to ensure successful completion. Once the concept has passed the initial feasibility, funding and accrual benchmarks, the process of study initiation including regulatory reviews (eg IRB, data safety and monitoring and scientific review, clinical research finance for budgetary review) are described. For novel drug trials, the process of an Investigational New Drug (IND) application with the Food and Drug Administration (FDA) is discussed.

3.1.3 |. Cooperative groups

Although the majority of clinical trials are investigator-initiated trials, a significant number of opportunities exist within the cooperative groups under the National Clinical Trials Network (NCTN) which includes the Alliance for Clinical Trials in Oncology, Southwest Oncology Group (SWOG), Children’s Oncology Group (COG), NRG Oncology and ECOG-ACRIN. In addition, the NCI Community Oncology Research Program (NCORP) includes 46 community sites with approximately 900 independent oncology practices with a large community-based program for cancer care delivery, which offers the opportunities for testing treatments in everyday care environments. A basic overview of the cooperative group structure is provided.

3.1.4 |. Responsibilities of the investigator and how to build a successful team

Many clinical trials in surgical oncology include a multi-specialty collaborative team including colleagues from medical oncology, radiation oncology, diagnostic and interventional radiology, pathology, and other clinical partners. Thus, early engagement in the design process ensures a diverse team with active collaboration. In addition, the incorporation of patient advocates and patient-reported outcomes in the design is of tremendous importance for the ability to accrue participants and remain patient-centered. Furthermore, inclusion of patient input also ensures the question is relevant and meaningful to patients and the trial design is sensitive and feasible to their needs.

Once the trial has passed the scientific and regulatory processes, the PI has a number of responsibilities to ensure the safety and integrity of the trial. Before activation, research team training including review of the protocol and communication with collaborators to clearly outline the roles and responsibilities of all team members. A site-initiation visit including overview of the trial objectives, eligibility criteria, schema and study visits are reviewed as well as plans for enrollment with accrual targets and screening strategies and data entry methods. Data integrity is of utmost importance and supervision of research staff and delegation of tasks are necessary to ensure high-quality data. Final regulatory considerations including FDA form 1572 (Statement of Investigator) and review of the delegation of authority log are reviewed. Additional financial considerations including insurance approval and sponsor invoicing are covered to ensure clear delineation of research vs clinical care costs. While these details were not expected to be learned on first pass, our goals in the course were to delineate the full spectrum of PI responsibilities to show that they can be systematically handled, especially with the right research support staff.

3.2 |. Small groups

The didactic courses provide the foundation for interactive small group sessions where participants discuss and refine their primary research question and hypothesis. In the first session, participants develop a PICO question of their choice based on their clinical and research interests. One month before the the first small group session, participants are asked to begin drafting the PICO question, as this allows for more time-effective refinement in the first session. The PICO questions are presented to their assigned small group by the trainee and then critiqued by two to three faculty as well as four to five other trainees in the same group. The goal by the end of the first small group session is to have a defined primary research question with an identified study population and appropriate primary endpoint, with an emphasis on feasibility. The feasibility issue is perhaps the most important realistic teaching point we convey since this is course is meant to help design each trainee’s first prospective study either as a fellow or as a junior faculty.

Between sessions, the fellows are encouraged to meet with faculty mentors to further refine their questions. In subsequent small group sessions, the proposal is expanded to define inclusion and exclusion criteria, primary and secondary endpoints, as well as very simple sample size estimates (to re-emphasize feasibility).

At the final session, attendees showcase their research questions in an auditorium to the entire group for feedback. Based on feedback from participants, this interaction is likely the most meaningful part of the course in that it opens the eyes of each trainee to many novel ideas and allows attendees to evaluate not only their own question but also learn how to analyze others’ designs. Statistical support is also available between sessions to refine the design.

4 |. POTENTIAL CHALLENGES

Like many things in surgical education, there is a consistent tension between clinical responsibilities and protected time for education/research, as both provide critical learning requirements for modern Surgical Oncology training programs. Important to our success was the strong faculty support in recognizing the value of this course given the departmental faculty’s own experience in clinical and translational research. With this, we were able to open the course to all fellows in Complex General Surgical Oncology, Breast Surgical Oncology, Endocrine Surgery, and Hepato-Biliary Surgery, as well as our T32 residents in Surgical Oncology. Additionally important was the dedicated commitment of the fellows to completing clinical tasks efficiently and allocating the time necessary between the course sessions to refine their trial concept. To respect the clinical service needs, we designed the course to be on three Friday afternoons, a time when traditionally there were fewer surgical cases and clinics scheduled. Furthermore, funding resources can also be a limiting factor. We were grateful to benefit from strong departmental leadership who provided resources to cover the costs of the sessions. All course faculty graciously donated their time which further helped to alleviate any financial burden.

As we look to expand the course to include a broader range of surgical subspecialties, there remain questions about scalability. On one hand, inclusion of additional trainees in clinical trial education is paramount, but this must be weighed against the ability to develop an interactive course where participants are comfortable actively engaging in discussions and sharing concepts that are works-in-progress. In general, faculty involvement at a ratio of 1:4 is ideal for maximal engagement, but this requires a number of faculty to dedicate a significant amount of time to remain actively involved.

5 |. COURSE EVALUATION

Throughout the years, we have requested feedback from the participating fellows in the form of post-course surveys. This has allowed us to refine the course content and presenting faculty. The response to the course has been overwhelmingly positive with this becoming a critical element of the fellows’ education and one which provides them an opportunity to incorporate research skills into their early career development.

6 |. FUTURE DIRECTIONS

As our understanding of clinical and translational cancer treatment strategies continues to change at an extraordinary pace, the need to be able to quickly and effectively identify novel effective treatments has become even more paramount. Surgeons (and especially surgical oncologists) play a central role in the multidisciplinary treatment of patients with early and advanced disease, and we are often the clinicians with primary access to tissue samples and the referral base to our medical and radiation oncology colleagues for adjuvant therapies. With this, the surgical oncologist has the unique opportunity and responsibility for clinical trial development in patients with earlier stages of disease, including novel multimodality therapeutic trials and perioperative pragmatic trials to improve the quality of life of our patients.