Jennifer M. Gierisch, PhD, MPH; Evan R. Myers, MD, MPH; Kristine M. Schmit, MD, MPH; Matthew J. Crowley, MD; Douglas C. McCrory, MD, MHS; Ranee Chatterjee, MD, MPH; Remy R. Coeytaux, MD, PhD; Amy Kendrick, RN, MSN; Gillian D. Sanders, PhD
This article was published online first at www.annals.org on 25 February 2014.
Disclaimer: The authors of this article are responsible for its content. Statements in this article should not be construed as endorsement by the U.S. Department of Veterans Affairs.
Acknowledgment: The authors thank Megan von Isenburg, MSLS, for help with the literature search and retrieval; Megan M. Chobot, MSLS, for project coordination; and Rebecca Gray, DPhil, for editorial assistance. They also gratefully acknowledge the contributions of members of the stakeholder panel (Peter Beitsch, MD; Laura Esserman, MD, MBA; Temeika Fairley, PhD; Brandel France de Bravo, MPH; Worta McCaskill-Stevens, MD, MS; Donna Pinto; Rinaa Punglia, MD, MPH; Joy Simha; and Debbie Saslow, PhD).
Grant Support: By a contract from PCORI.
Potential Conflicts of Interest: Disclosures can be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M13-2548.
Requests for Single Reprints: Jennifer M. Gierisch, PhD, MPH, Department of Medicine, Duke University Medical Center, Campus Box 104427 DUMC, Durham, NC 27701; e-mail, firstname.lastname@example.org.
Current Author Addresses: Dr. Gierisch: Department of Medicine, Duke University Medical Center, Campus Box 104427 DUMC, Durham, NC 27701.
Dr. Myers: Duke University Medical Center, Box 3279, 244 Baker House, Durham, NC 27710.
Drs. Schmit, McCrory, Coeytaux, and Sanders and Ms. Kendrick: Duke Clinical Research Institute, PO Box 17969, Durham, NC 27715.
Dr. Crowley: Durham Veterans Affairs Medical Center, Health Services Research & Development Service (152), 508 Fulton Street, Durham, NC 27705.
Dr. Chatterjee: Sutton Station Internal Medicine, 5832 Fayetteville Road, Suite 113, Durham, NC 27713.
Author Contributions: Conception and design: J.M. Gierisch, D.C. McCrory, A. Kendrick, G.D. Sanders.
Analysis and interpretation of the data: J.M. Gierisch, E.R. Myers, R. Chatterjee, R.R. Coeytaux, G.D. Sanders.
Drafting of the article: J.M. Gierisch, E.R. Myers, G.D. Sanders.
Critical revision of the article for important intellectual content: J.M. Gierisch, E.R. Myers, M.J. Crowley, D.C. McCrory, G.D. Sanders.
Final approval of the article: J.M. Gierisch, E.R. Myers, M.J. Crowley, D.C. McCrory, R. Chatterjee, R.R. Coeytaux, A. Kendrick, G.D. Sanders.
Provision of study materials or patients: G.D. Sanders.
Statistical expertise: J.M. Gierisch.
Obtaining of funding: J.M. Gierisch, G.D. Sanders.
Administrative, technical, or logistic support: A. Kendrick.
Collection and assembly of data: J.M. Gierisch, E.R. Myers, K.M. Schmit, M.J. Crowley, R. Chatterjee, R.R. Coeytaux, A. Kendrick, G.D. Sanders.
Gierisch JM, Myers ER, Schmit KM, Crowley MJ, McCrory DC, Chatterjee R, et al. Prioritization of Research Addressing Management Strategies for Ductal Carcinoma In Situ. Ann Intern Med. 2014;160:484-491. doi: 10.7326/M13-2548
Download citation file:
Published: Ann Intern Med. 2014;160(7):484-491.
Ductal carcinoma in situ is a common finding in women having mammography screening, and there is considerable uncertainty about the balance of harms and benefits of different management options. This article outlines the process for developing a prioritized research agenda for the Patient-Centered Outcomes Research Institute as informed by a diverse group of stakeholders on the management of ductal carcinoma in situ. Evidence gaps were identified by reviewing existing literature and engaging diverse stakeholders to refine these gaps. Stakeholders ranked evidence gaps by importance from their perspectives using a forced-ranking prioritization method. PubMed was searched for relevant recent studies, and ClinicalTrials.gov was searched for relevant ongoing trials for the 10 highest-ranked evidence gaps. Strengths and limitations of different study designs were assessed to address gaps. Stakeholders prioritized evidence gaps related to incorporation of patient-centered outcomes into future research, development of better methods to predict risk for invasive cancer, evaluation of a strategy of active surveillance, and testing of decision-making tools. The degree to which prioritized evidence gaps may have already been addressed is uncertain because a comprehensive systematic review has not been done.
Ductal carcinoma in situ (DCIS) is a condition in which abnormal cells line the milk ducts of the breast but have not invaded the underlying breast tissue. Approximately 55 000 new cases of DCIS were expected to occur among U.S. women in 2013, which accounted for approximately 24% of all new breast “cancer” diagnoses (1). Although the annual incidence of DCIS is less than that of invasive breast cancer, expanded use of screening mammography has led to a dramatic increase in the diagnosis of DCIS (32.5 cases per 100 000 women) (2). There is considerable uncertainty about the optimal clinical management of DCIS because of the lack of reliable methods for distinguishing DCIS that would never become symptomatic from DCIS that is likely to progress to life-threatening invasive cancer (3).
Current treatment options for DCIS include mastectomy, breast-conserving surgery (BCS), and BCS plus radiation therapy (2, 4–6). There are limited randomized, controlled trials (RCTs) on which to base decisions about the tradeoffs among DCIS treatment options, and the existing trials may not capture all relevant patient-centered outcomes. For example, 4 large RCTs comparing BCS alone with BCS plus radiation therapy all show similar, very high overall survival regardless of treatment (7–10). The addition of radiation therapy significantly reduces the risk for local recurrence; approximately one half of all recurrences are DCIS rather than invasive cancer (5, 11). The long-term benefit of preventing the recurrence of an asymptomatic condition that may or may not progress to cancer is difficult to quantify. In other conditions where screening may detect early cancer or cancer precursors that may not progress to a life-threatening stage, such as early-stage prostate cancer, active surveillance or watchful waiting (an observation strategy) is an acceptable option for some patients. A similar strategy may be acceptable for some women diagnosed with DCIS, although no studies have compared management strategies that involve no immediate treatment (that is, observation or active surveillance) versus immediate treatments.
The most important potential benefit of treatment of DCIS is the reduction in the number of deaths due to invasive breast cancer. For many women with a DCIS diagnosis, this may be an unrealized benefit. Some proportion of DCIS may never progress to life-threatening invasive cancer, so these women only have the adverse outcomes of diagnosis and treatments. Both active management strategies and surveillance approaches present tradeoffs relevant to patient-centered outcomes, such as symptoms, function, and well-being. Given the clinical importance of DCIS, the diversity of available treatment strategies, variety of patient-centered outcomes of interest, and areas of uncertainty, we sought to create a prioritized research agenda for the Patient-Centered Outcomes Research Institute (PCORI) that would incorporate different stakeholders' perspectives.
A central goal of PCORI is to engage stakeholders in its work in a meaningful way. Between September 2012 and January 2013, PCORI undertook a broad effort tosolicit research topics to consider for targeted funding from patients, caregivers, researchers, and results of previous prioritization processes by groups, such as the Institute of Medicine. In April 2013, PCORI's Assessment of Prevention, Diagnosis, and Treatment Options program, together with the program's external advisory panel, have identified DCIS management as an important topic with unmet research needs. Then, PCORI tasked the Duke University Evidence Synthesis Group (ESG) to work with various stakeholders to identify and prioritize the future research that is most needed by patients and other decision makers on this topic.
Our approach to prioritizing future research and developing recommendations for targeted future funding by PCORI included several steps (Figure 1). These included appraisals of recent systematic reviews to preliminarily identify important evidence gaps, transformation of evidence gaps into research questions and organization of the questions within a comprehensive analytic framework, engagement of stakeholders to identify additional gaps and prioritize research needs or questions, scans of recently published and ongoing studies relevant to the stakeholders' list of prioritized research needs, and identification of potential study designs to address prioritized research questions.
Overview of prioritization process
Adapted from reference 12. ESG = Evidence Synthesis Group; PCORI = Patient-Centered Outcomes Research Institute.
We used an iterative process to identify evidence gaps for DCIS management. First, we identified and appraised recent published systematic reviews, clinical practice guidelines, and future research needs documents to develop an initial list of evidence gaps. This list was neither exhaustive nor prioritized. Next, we organized these gaps according to the populations, interventions, comparators, outcomes, timing, and setting of interest and transformed them into research questions informed by the criteria.
We convened a group of 9 stakeholders, including clinical experts and researchers in DCIS management strategies, representatives from federal and nongovernmental funding agencies, representatives from relevant professional societies, health care decision makers and policymakers, and representatives from related consumer and patient advocacy groups (Table 1 of the Supplement). Within each of these categories, we sought to identify a person who was either familiar with the clinical area and its current uncertainties or brought a specific methodological expertise to the stakeholder panel. The University of Minnesota Evidence-based Practice Center, which did a comparative effectiveness review on DCIS management in 2009 (13), and PCORI's patient engagement group provided input on potential stakeholders. Each potential stakeholder completed a disclosure statement and was screened for conflicts of interest. We solicited and received stakeholder input at various points in the process through group discussions via teleconferences detailing the process and outlining existing evidence gaps, Web-based surveys to obtain priority ranking of topics, and e-mail communications to solicit or provide further clarification of research gaps from teleconferences.
On the basis of the expanded list of 30 evidence gaps and informed by an analytic framework in a recent evidence synthesis on DCIS (13), we constructed an analytic framework that depicted the identified knowledge gaps within the context of the populations, interventions, comparators, outcomes, timing, and setting criteria (Figure 2). This framework shows that women diagnosed with DCIS may have pretreatment evaluations. After these evaluations, women may undergo several management strategies. These management strategies affect patient-centered outcomes (such as development of comorbid conditions, DCIS recurrence, invasive cancer, and sexual functioning). Throughout the analytic framework, population factors (such as socioeconomic, family history, and genetic factors) and effects of decisional uncertainty about management strategies were highlighted as potential modifiers.
DCIS = ductal carcinoma in situ; FRN = future research need; MRI = magnetic resonance imaging.
After expansion of the identified research priorities, stakeholders were invited to rank the revised future research needs online. The survey used a forced-ranking prioritization method described by the Agency for Healthcare Research and Quality Evidence-based Practice Center's Future Research Needs projects (14), whereby participants were given 10 votes that could be allocated to any identified research priorities, with a maximum of 3 votes per item. The stakeholders were not given specific prioritization criteria to use but rather were told to decide, on the basis of their perspectives, which were the most important unanswered research questions in DCIS management. Only the priorities in the top tier (n = 10) moved on to the final stage of horizon scan and study design determination. Stakeholders were informed of the final ranking of future research priorities.
We performed 2 database searches to identify recently published and ongoing studies relevant to the top-tier future research questions resulting from the stakeholder forced-ranking prioritization exercise. We searched PubMed to identify relevant literature published since 2008 and ClinicalTrials.gov for ongoing studies.
Members of the ESG team reviewed the titles and abstracts identified by searching PubMed for applicability to the top-tier research questions; full texts were then obtained and screened for all citations passing the title-and-abstract screening. At both the title-and-abstract and full-text screening stages, articles were included if they met all of the following criteria: presented original data or secondary analysis of data from an RCT, prospective or retrospective observational study, or relevant modeling study; included data for a pretreatment evaluation or management strategy of interest; included women with DCIS; and had a stated objective that could be categorized according to our identified list of research priorities.
We searched ClinicalTrials.gov using the keywords “ductal carcinoma in situ” and “DCIS.” A member of the ESG team reviewed all study abstracts resulting from the search and coded them as potentially relevant to 1 or more of the identified research priorities. We then abstracted study type (such as observational or RCT), recruitment status, and sample size.
We considered the advantages and disadvantages of various potential study designs for each top-tier future research priority. We adapted a conceptual framework (Figure of the Supplement) for recommending study designs on the basis of our previous future research needs report (15). In brief, our approach to recommending study designs for specific research priorities was to emphasize the study design with the least risk of bias but greatest likelihood of completion. For example, if an RCT is not feasible, then a meta-analysis of RCTs may be most appropriate and feasible. If meta-analysis of RCTs is not feasible, then meta-analysis of observational studies could provide valuable information to address the research priority. This approach considers the feasibility and validity of the study to answer the question and assumes that the question is not answered by current research. We cannot determine with certainty the degree to which those evidence gaps have already been addressed because a comprehensive systematic review has not been done for many of these research questions. Senior ESG investigators completed research design considerations; we did not consult with stakeholders on this step.
This work was funded by PCORI. The funding source stipulated the topic for prioritization but did not participate in the literature search, determination of study eligibility criteria, data analysis or interpretation, or preparation or approval of the manuscript for publication. The funding source reviewed a draft version of the manuscript and provided suggestions to clarify language describing the process in selecting this topic for prioritization.
The panel of 9 stakeholders expanded the initial list of 20 evidence gaps to 30 and identified several common themes relevant to the evidence gaps. The overarching theme behind most comments was that there is considerable uncertainty about diagnosis (that is, whether the condition is truly limited to DCIS or whether there are areas of invasive cancer), prognosis (that is, the likelihood that a given DCIS lesion in a given patient will progress to life-threatening invasive cancer if left untreated), and treatment (that is, the likelihood that a given treatment option will prevent the development of life-threatening invasive cancer, whether that likelihood will change based on specific patient or tumor characteristics, and whether the reduction in risk is enough to justify the potential undesirable outcomes of that treatment). Given this uncertainty, stakeholders emphasized the need for comparative studies of techniques for improving diagnostic certainty (that is, the comparative sensitivity and specificity of different imaging methods to guide pathologic diagnosis and surgical planning) and prognostic certainty, particularly for identifying DCIS lesions that are least likely to progress to life-threatening invasive cancer. Stakeholders also stated that there is a need to assess decision-support interventions to manage uncertainty about the short- and long-term implications of a DCIS diagnosis, how DCIS differs from invasive breast cancer, and the relative harms and benefits of different management options. There was particular interest in evaluating strategies, such as peer-to-peer support or lifestyle modifications.
Several stakeholders felt strongly that uncertainty about the long-term effect of DCIS on breast cancer mortality rates and the high survival rates seen with all current treatments justified studies comparing outcomes achieved with an active surveillance strategy versus more conventional management options. Such studies would be facilitated by the availability of validated tools to estimate risk for progression to invasive cancer. All stakeholders agreed on the need to incorporate a broader range of patient-centered outcomes into studies of DCIS diagnosis and management.
The Appendix Table shows the 30 final potential research topics and stakeholder ranking. The top 10 future research needs prioritized by stakeholders were related to developing risk-stratification models, comparing a management strategy of no immediate treatment (that is, observation or active surveillance) versus immediate treatment, understanding the influence of clinical or biological characteristics on DCIS management strategies, assessing the effect of different approaches to communicating the diagnosis of DCIS to patients, testing decision-making tools, assessing preoperative imaging, examining the effect of management strategies on comorbid conditions, optimizing radiation therapy, identifying key patient-centered outcomes, and assessing the effect of management strategies on invasive cancer.
Appendix Table. Final Ranking of Future Research Needs for Management Strategies for DCIS
We conducted an exploratory analysis of stakeholder prioritization by stakeholder role (that is, patient advocacy, researchers, and policymakers). Each stakeholder group ranked individual future research needs differently, but 4 research needs were highly ranked by all stakeholder groups: developing risk-stratification models, comparing the management strategy of no immediate treatment versus immediate treatment, assessing different approaches to communicating DCIS diagnosis, and testing decision-making tools. The theme of uncertainty persisted across the individual rankings and was manifested in slightly different ways. The patient advocates tended to highly rank research into ways of minimizing uncertainty, whereas researchers put a high priority on managing uncertainty; because local variation presumably arises when there is substantial uncertainty, policymakers prioritized quantifying uncertainty. It is important to note that the stakeholder panel was a small, nonrepresentative sample, and many of the stakeholders represented multiple perspectives. Thus, these findings may not be generalizable to stakeholder groups as a whole.
The horizon scan demonstrated a high-level match between overall stakeholder priorities and recent or ongoing research. Seven of the top 10 prioritized research needs had more than a dozen recent or ongoing studies. The prioritized areas with the most identified studies included the understanding of the influence of clinical or biological characteristics on DCIS management strategies (n = 35), optimization of radiation therapy (n = 33), and effect of DCIS management on invasive cancer (n = 55). When rank ordering was taken into account, 3 of the 5 highest-ranked research areas (management strategy involving no immediate treatment vs. immediate treatment, different approaches to communicating the DCIS diagnosis to the patient, and decision-making tools compared with usual care) had few recently completed or ongoing studies.
Our PubMed search identified 1417 articles. Of these, 105 met our inclusion criteria and included 15 systematic reviews, 3 RCTs, 83 cohort studies, 1 case–control study, and 3 model-based studies. Sample sizes ranged from 11 to 23 547. Only 12 studies were active comparator studies, 50 studies either were placebo-controlled or used standard of care as the comparison, and 43 studies had no comparator. No recently published studies pertained to 2 of the prioritized research areas, including decision-making tools compared with usual care and identification of the most important patient-centered outcomes. Only 1 recently published study addressed the priority area of different approaches to communicating the DCIS diagnosis to the patient.
Our search of ClinicalTrials.gov yielded 206 protocols submitted since 1 January 1998. Of these, 79 were open and enrolling, 43 were active and not enrolling, 3 were enrolling by invitation only, 58 had been completed, and the remaining 23 were categorized as unknown status. We identified 37 protocols as potentially relevant to the top-tier research questions. These protocols were a mix of study designs: 12 RCTs, 5 observational studies, and 20 nonrandomized, interventional trials. Sample sizes ranged from 2 to 4300 patients.
Several of the prioritized research questions had no identified ongoing studies. These included development and validation of risk-stratification models; management strategies involving no immediate treatment (that is, observation or active surveillance) versus immediate treatment with surgery, radiation, or medical therapy; different approaches to communicating the DCIS diagnosis to the patient; and identification of the most important patient-centered outcomes. Tables 2 to 11 of the Supplement detail key characteristics of the included PubMed and ClinicalTrials.gov articles separately for each of the top-tier future research needs.
For each top-tier future research priority, we considered advantages and disadvantages of various potential study designs. The Table presents our summary suggestions for appropriate study designs to address the top-tier research priorities. Table 12 of the Supplement provides more thorough details.
Table. Study Designs Suggested for Top-Tier Future Research Needs for Management Strategies for DCIS
This article outlines our process for developing a prioritized research agenda for PCORI as informed by a diverse group of stakeholders. We developed a list of 30 potential future research topics in DCIS management on the basis of the existing literature and with input from stakeholders. The stakeholders prioritized these topics through a forced-ranking process. We then examined recently published and ongoing studies to identify research relevant to the top 10 future research priorities to assist PCORI in developing future targeted funding opportunities.
Stakeholder input on research priorities was consistent with other recent reviews and recommendations. The 2009 National Institutes of Health State-of-the-Science Conference statement (3) ranked 6 critical research questions for DCIS diagnosis and management. Only 1 of these (that is, creation of a database that collects clinical, pathologic, and radiologic characteristics of DCIS) does not overlap with the priorities identified by our stakeholder panel. In particular, both groups suggested that consideration be given to renaming DCIS to avoid the use of the word “carcinoma” because use of this term may contribute to patient overestimation of long-term risk, increased urgency about making treatment decisions, and subsequent choice of overly aggressive management options. Both groups highly ranked the validation of risk-stratification models to inform treatment choice. Likewise, our stakeholder panel ranked comparative research into methods for communicating information about the diagnosis and prognosis of DCIS to patients as one of the most important future research needs. Our process also identified unique research needs related to the most important patient-centered outcomes and framed them from the perspective of a patient rather than a health care delivery process (that is, different approaches to communicating DCIS to a patient compared with using standard DCIS terminology). The recent Institute of Medicine report about delivering high-quality cancer care (16) also shares some recommendations with the research priorities we identified here but to a much lesser extent than the National Institutes of Health State-of-the-Science Conference statement. Parity exists on issues pertaining to use of decision aids, improvement of patient–provider communications, and importance of collecting patient-reported outcomes. However, the Institute of Medicine report also made recommendations about training of cancer care staff, use of information technology, and transparency of cancer care cost that did not make the list of 30 research gaps.
The overarching theme that emerges from the deliberative process described here is that uncertainty about the long-term implications of a diagnosis of DCIS is the major factor affecting the ability of patients to make decisions about management options that reflect their preferences. Two of the 3 highest-ranked questions focus on research to help generate estimates of the long-term risk for a given patient whose DCIS progresses to life-threatening cancer, using available data on patient and disease characteristics (including imaging, pathology, and biomarker testing), and perhaps integrating these into a multivariate risk-prediction model. These models are used to help identify women at high risk for invasive breast cancer and help inform decisions about primary and secondary prevention strategies in such patients. Both the literature search and our review of ongoing research in ClinicalTrials.gov show a substantial amount of research focused on identifying specific predictors but relatively little research on integrating this work into a reliable, valid, and feasible tool to help with clinical decision making. Stakeholders also prioritized research into better strategies for helping patients cope with the current uncertainty of a DCIS diagnosis. This is reflected by the high rankings for research of tools and methods for helping patients make decisions in the face of uncertainty and the optimal method for communicating diagnostic and prognostic information to patients. In our stakeholder discussions, patient representatives were vocal in identifying uncertainty both before and after treatment as a major cause of ongoing distress, a finding that is supported by the literature.
Another contributor to patient uncertainty may be that even when there is relatively high-quality evidence comparing outcomes of different management options, the outcomes reported may not include those important to a specific patient or their clinical relevance may be unclear. For example, high-quality evidence shows that the addition of radiation therapy to BCS reduces the risk for local recurrence compared with BCS alone, but many of these local recurrences are recurrences of DCIS. Whether the direct harmful effects of adding radiation are balanced by the benefits of preventing the reappearance of an asymptomatic condition with uncertain implications for long-term illness and death is not clear. Research aimed at identifying which outcomes are most important to patients for making decisions would help guide future clinical research as well as the development of future guidelines and quality measures.
Our prioritization process is not without limitations. Although we took great efforts to be comprehensive, it is possible that the list of future research needs we generated and expanded with stakeholder feedback does not reflect the full range of possible future research. In addition, we engaged a relatively small number of stakeholders. It is also possible that another group of stakeholders might rank the identified future research needs differently. Still, we included a diverse stakeholder panel with a range of expertise in determining these priorities with a particular focus on patient-centered research. Also, because a comprehensive systematic review has not been done for many of the identified evidence gaps, we cannot determine with certainty the degree to which prioritized future research needs have already been addressed.
As long as mammography continues to be the primary tool available to reduce breast cancer mortality rates, many women will be given a DCIS diagnosis during their lifetime. Thus, improvement of the ability of women given a DCIS diagnosis to make informed decisions about management options will be a critical need for the foreseeable future. A workgroup of 9 stakeholders representing diverse perspectives identified 10 research areas as the highest priority for future research for patient-centered management strategies for DCIS that, if studied, have the potential to resolve some of the uncertainty surrounding a DCIS diagnosis.
The In the Clinic® slide sets are owned and copyrighted by the American College of Physicians (ACP). All text, graphics, trademarks, and other intellectual property incorporated into the slide sets remain the sole and exclusive property of the ACP. The slide sets may be used only by the person who downloads or purchases them and only for the purpose of presenting them during not-for-profit educational activities. Users may incorporate the entire slide set or selected individual slides into their own teaching presentations but may not alter the content of the slides in any way or remove the ACP copyright notice. Users may make print copies for use as hand-outs for the audience the user is personally addressing but may not otherwise reproduce or distribute the slides by any means or media, including but not limited to sending them as e-mail attachments, posting them on Internet or Intranet sites, publishing them in meeting proceedings, or making them available for sale or distribution in any unauthorized form, without the express written permission of the ACP. Unauthorized use of the In the Clinic slide sets will constitute copyright infringement.
Hematology/Oncology, Research and Reporting Methods, Breast Cancer.
Results provided by:
Copyright © 2016 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use
This PDF is available to Subscribers Only