Challenges in DCIS Risk Communication and Decision-Making: Report from an American Cancer Society and National Cancer Institute Workshop

Abstract

In September 2010, the American Cancer Society and National Cancer Institute convened a conference to review current issues in DCIS risk communication and decision-making and to identify directions for future research. Specific topics included patient and healthcare provider knowledge and attitudes about DCIS and its treatment, how to explain DCIS to patients given the heterogeneity of the disease, consideration of nomenclature changes, and the utility of decision tools/aids. This report describes the proceedings of the workshop in the context of the current literature and discusses future directions. Evidence suggests that there is lack of clarity about the implications and risks of a diagnosis of DCIS, among patients, providers and researchers. Research is needed to understand better the biology and mechanisms of the progression of DCIS to invasive breast cancer and the factors which predict those subtypes of DCIS which do not progress, as well as efforts to improve communication and informed decision-making surrounding DCIS.

Introduction and Background

At the National Institutes of Health (NIH)State-of-the-Science conference on ductal carcinoma in situ (DCIS) held in September 2009, it became evident that there were a number of issues related to communication of risk with patients regarding the disease.^1–4 Recommendations were made for more research to be done in the area of DCIS risk communication and for the development of decision aids and for strategies to integrate them into clinical practice. The Consensus Panel’s Statement also concluded: “because of the noninvasive nature of DCIS, coupled with its favorable prognosis, strong consideration should be given to remove the anxiety-producing term ‘carcinoma’ from the description of DCIS.”²

In consideration of these issues, the American Cancer Society (ACS) and National Cancer Institute (NCI) convened a workshop in September 2010 to review available evidence and issues regarding DCIS risk communication and decision-making and to identify directions for future research. Invited participants included a small group of clinical and basic scientists, advocates, and communication specialists (see Appendix A for list of participants). Patient-provider communication and informed medical decision-making surrounding DCIS diagnosis and treatment, psychosocial outcomes of women with DCIS, and consideration of changing the nomenclature were addressed. The primary goal of the ACS/NCI workshop was to review what is known about these issues and to develop recommendations and strategies to improve them. A secondary goal was to discuss what we know about the association between DCIS nomenclature and distress or confusion, and the pros, cons, and feasibility of changing the nomenclature for DCIS and to identify areas where more information is needed.

Conference participants heard presentations by experts, followed by question and answer sessions and group discussions. The speakers included: Worta McCaskill –Stevens, MD, MS, Ann Partridge, MD, MPH, Joann Elmore, MD, MPH, Karen Sepucha, PhD, Stuart J. Schnitt, MD, Fattaneh.A. Tavassoli, MD, Umberto Veronesi, MD, Neeraj Arora, PhD, and Mary Lou Smith, JD. This article summarizes the key points from presentations and discussions in the context of the current literature from the ACS/NCI workshop as well as future directions to address the issues raised.

A brief overview of the NIH State-of- the-Science conference was presented to set the stage for the discussion.^{1, 2} For purposes of discussion, the following definition of DCIS was used: DCIS is the replacement of normal ductal cells with a spectrum of abnormal cells confined to the breast ducts. The diagnosis has increased dramatically in the era of mammographic screening such that DCIS is now diagnosed in approximately 50,000 women in the U.S. alone annually.^5–7 Research has revealed that DCIS is a term that encompasses a heterogeneous group of lesions with a variable natural history and risk of progression to invasive breast cancer.^8–11 The natural history and risk of progression to invasive disease has been studied in women who were found to have DCIS on retrospective review of breast biopsies originally categorized as benign and who, therefore, underwent no more than a diagnostic biopsy. In these studies which included primarily cases of low grade DCIS, up to 40% of women were diagnosed with invasive cancer in the ipsilateral breast with follow-up of more than 30 years.¹² Data such as these have historically been used to justify aggressive local therapy for the disease.¹³ Current standard treatments options for DCIS include excision followed by radiation, wide excision alone, mastectomy, and tamoxifen after excision, with or without radiation. Among women treated with breast-conserving methods, there remains a broad range of local recurrence rates, though it is generally lower than in the setting of invasive disease. Of note, approximately 50% of local recurrences following breast conserving treatment will be invasive cancers and 50% DCIS (Figure 1).

Figure 1.

Risk of recurrence after DCIS (adapted from Burstein et al.,¹³ by Thea Tlsty, 2010)

Risk stratification among women with DCIS has been an area of active research for more than two decades but remains challenging. There are currently no clinical factors, histopathologic features, or molecular markers singly or in combination that permit reliable stratification of risk for either invasive or non-invasive recurrence for individual patients. The role of molecular markers and gene expression signatures to identify patients at risk for future events of DCIS and invasive breast cancer is evolving, but the clinical utility of these tests is at the present time uncertain.^14–16 Consequently, there is controversy about the optimal treatments for women with DCIS.¹³ Overall, however, women with DCIS have a very favorable prognosis and a diagnosis of DCIS is not likely to affect a woman’s survival. Further, there has been concern about the overtreatment of DCIS, particularly small lesions that might not have ever become clinically evident.¹⁷ Thus, addressing the problem of suboptimal communication and the potential for poor quality decision-making and psychosocial outcomes is of great clinical importance.

Knowledge and Communication about DCIS

Evidence suggests that DCIS is not a disease with which most women are familiar. In a cross sectional survey of 479 U.S. women in 1997, only 6% reported that they had heard of DCIS and only 7% agreed that there are “some types of breast cancer that grow so slowly that even without treatment they would not affect a woman’s health”.¹⁸ Among women who are diagnosed with DCIS, there is a lack of understanding of the disease entity, particularly with regard to the non-invasive nature and whether or not it is “cancer” or could spread to other places in a woman’s body and become life-threatening.^19–24 For example, in a letter to the BMJ, a patient with DCIS understandably bemoaned the fact that during one appointment with her physician, she was told both that she did have cancer and that she did not have cancer.²⁰ Women’s confusion is potentially compounded by the use of the term “carcinoma” as this implies for many women that they have invasive breast cancer. In addition, treatments recommended for DCIS such as mastectomy, partial mastectomy followed by radiation, and hormonal therapy are also often recommended for women with invasive disease, possibly leading women to think of DCIS as the same as invasive cancer.^{21, 25, 26} However, there are no rigorous data available on the reaction of women to the use of the term “carcinoma” or to the alternative terms that have been proposed. Further, while DCIS is classified as Stage 0 breast cancer according to the AJCC, the common use of the terminology “breast cancer” to refer to both DCIS as well as invasive disease likely adds to the confusion given the different risks associated with invasive compared to non-invasive breast cancer.^{27, 28}

There is very little available information regarding physicians’ perceptions and communications strategies in caring for women with DCIS despite the substantial evidence that management of DCIS is strongly related to physician recommendations and varies substantially nationally and internationally.^28–36 A cross-sectional survey of 151 U.S. physicians who care for women with DCIS revealed heterogeneity among them regarding terms used to describe DCIS when speaking with patients and management approaches for the disease.²⁸ The majority of these physicians rated the emotional distress that women generally experience when diagnosed with DCIS as high and perceived the treatment decision-making process to be quite difficult for these women. Most (78%) also indicated that the DCIS decision-making process was as difficult (36%) or more difficult (42%) than that for women with invasive breast cancer. Finally, only 63% indicated that a diagnosis of DCIS posed little or no risk to a woman’s overall long-term health. A survey of 296 health professionals in the United Kingdom involved with the treatment of DCIS patients confirmed diverse perceptions of the disease, difficulty explaining DCIS to patients, and heterogeneity in terminology used.³⁴ It is likely that the clinical heterogeneity, uncertainty about the natural history of DCIS, particularly for any given woman, as well as the controversies surrounding optimal treatment, contribute to the heterogeneous management approaches and likely some physician discomfort with the disease. Further evaluation of the effects of physician attitudes, communication strategies and management approaches for women with DCIS on patient outcomes is clearly needed.

Risk Perceptions, Anxiety and Quality of Life in Women with DCIS

Given the confusion among patients about the entity of DCIS, and heterogeneous views among providers, it is not surprising that many women with DCIS are anxious about their disease and overestimate the risks they face.^{23, 37–47} In a recent cross-sectional survey of 144 women diagnosed with DCIS in Australia, many women expressed both misunderstanding and confusion about DCIS and the associated risks, and desired more information about their breast disease.²³ In this study, 73% of women described their disease as early stage breast cancer, and only 19% of participants were aware that not all women with DCIS will develop invasive breast cancer. Sixty percent of women thought DCIS can metastasize and 27% were unsure about this. Further, approximately half expressed high decisional conflict when considering treatment options.

In a large prospective cohort study of U.S. women with newly diagnosed DCIS (N=487), a substantial proportion of participants harbored inaccurate perceptions about the breast cancer risks they faced including both local and distant recurrence. For example, approximately 25% of women perceived at least a moderate risk of DCIS spreading to other parts of their bodies at baseline, 18-month and 5-year follow-up.^{39, 47} Increased anxiety was significantly associated with inaccurate risk perceptions.

Several studies have also found that women with DCIS have similar risk perceptions and anxiety compared to women with invasive breast cancer.^41–44 In a cross-sectional study of 228 women with either DCIS or invasive disease, women with DCIS perceived that they had essentially the same risks of local recurrence, distant recurrence and death compared to women with invasive cancer.⁴¹ In a prospective study of 549 women with newly diagnosed early-stage breast cancer including a substantial proportion with only DCIS (34%), patients who were white (OR = 5.88, 95% CI 3.39–10.19) and had greater state anxiety (OR = 1.04, 95% CI 1.02–1.07) were more likely to report higher risk of recurrence, while patients who received radiotherapy (OR = 0.72, 95% CI 0.54–0.96) and had more social support (OR = 0.59, 95% CI 0.46–0.75) were less likely to report higher risk of recurrence. Cancer stage was not significantly associated with perceived risk of recurrence and perceived risk of recurrence did not change significantly over time.⁴⁴

In the only published study with a pre-morbid assessment of health-related quality-of-life (HRQoL), Nekhlyudov and colleagues compared changes among women who developed DCIS compared to those who did not in two Nurses' Health Study cohorts using the Medical Outcomes Survey Short-Form 36 health survey.³⁸ Women who were diagnosed with DCIS had small, but statistically significantly greater declines in the domains of role limitations due to physical problems, vitality, and social functioning than women without DCIS. Among those with DCIS, clinically significant declines were more often observed within 6 months of the diagnosis in the domains of social functioning and mental health than after 6 months from diagnosis.³⁸ These data are consistent with other studies which suggest that despite increased anxiety and inaccurate risk perceptions among many women with DCIS, the effects of the diagnosis and treatment on overall HRQoL appear to be limited.^{28, 37, 42, 47}

Inaccurate, heightened perceptions of breast cancer risks among women with DCIS have been associated repeatedly with increased anxiety. It is not clear from the current literature whether women with high baseline anxiety are more likely to perceive their risks inaccurately upon the diagnosis of DCIS or whether inaccurate risk perceptions are driving up anxiety levels. Regardless, it is likely that any intervention to improve risk perceptions will need to address not only understanding and informational gaps, but address and manage anxiety.

Decision-Making in DCIS

Inaccurate risk perceptions and anxiety about DCIS may hamper optimal, high quality, shared decision-making. Patient centered care entails shared decision-making between patients and providers and requires that the patient is engaged and accurately informed about options and outcomes so that treatment decisions can be consistent with the patient’s goals, preferences, and values.^48–52 Patient-centered care not only incorporates the patient (and potentially, loved ones) perspective in the care planning and delivery, but aims to provide ongoing support to meet patient needs (medical and psychosocial) as best as possible and implies responsiveness to those needs. This requires patient-centered communications which includes fostering healing relationships with providers in which trust is key, accurate information exchange regarding implications of disease and potential risks and benefits of treatments, provider response to emotions, assistance with decision-making and managing uncertainty as well as enabling self-management.⁵³ Providing this optimal care in our complex healthcare environment for every patient poses challenges and research has suggested that substantial gaps exist.⁵⁰ In cancer survivors, there is evidence that physicians who adopt a participatory decision-making style are likely to facilitate patient empowerment and enhance patients' HRQoL.⁵⁴

Optimizing patient-centered care may be particularly valuable when caring for patients with DCIS where there is such confusion regarding the diagnosis as well as uncertainty in available knowledge about the disease. Intervention directed toward improving communication styles among physicians who care for women with DCIS may lead to more accurate risk perceptions, more informed decision-making and better psychosocial outcomes in this population although this has not been studied prospectively. Among women with DCIS, large cross-sectional, population-based studies have revealed that many women do not perceive that they were offered a choice between surgical treatment options.^{29, 32} Not surprisingly, surgeon recommendations, which appear to take into account important clinical factors, heavily influence treatment decisions. Patient attitudes also appear to play an important role in treatment decisions. Knowledge about differences in clinical benefits and risks between surgery options has been found to be low among patients and satisfaction with the decision-making process significantly lower in women who did not perceive a choice between surgery options.²⁹

There is also some evidence that attention to distress, as well as informational needs in women with DCIS, may improve psychosocial outcomes. A small cross-sectional interview study of a multiethnic group of women with DCIS revealed ethnic differences in cognitive and emotional responses to DCIS.⁵⁵ White women generally reported a better understanding of their diagnosis and treatment, and Latinas generally reported more distress. Regardless of ethnicity, the women preferred that physicians discuss DCIS treatment options and attend to their informational and emotional needs. Further, satisfaction was associated with adequate information, expediency of care, and physicians' sensitivity to patients' emotional needs.

Patient decision aids may help to improve risk communication, decision-making, and distress for women with DCIS. They have been shown to be feasible and acceptable, increase patient involvement, and are more likely to lead to informed values-based health-related decisions.⁵⁶ They also help patients to make health decisions and reduce decisional conflict. Further, decisions made with the use of decision aids are more likely to be based on better knowledge, more realistic expectations, clearer values, and better communication. They have been studied among women with invasive breast cancer and found to improve communication and knowledge, reduce decisional conflict, and enable women to make a choice regarding surgical treatments.^{57, 58} No published studies of a decision aid have focused on women with DCIS, although there are available decision aids in use currently focused on treatment decisions for women with DCIS. (http://decisionaid.ohri.ca/Azsumm.php?ID=1187, accessed January 5, 2012)

There is clearly a need to identify the mediators and moderators of the link between communication and patient outcomes in women with DCIS. Future research is warranted to understand and intervene in the complex relationship between risk perceptions, anxiety/distress and decision-making. Decision aids are important tools to facilitate ongoing patient-clinician communication (not replace it) and further research on how decision aids effect communications, decision-making, knowledge and risk perceptions as well as psychosocial outcomes among women with DCIS should be conducted.

Nomenclature Issues: What’s in a Name?

There has been discussion over the past few decades about modifying the nomenclature of DCIS to remove the term “carcinoma.” In particular, proponents of this approach have recommended replacing DCIS with “Ductal Intraepithelial Neoplasia” or “DIN” terminology.^59–61A new clinical and biological TNM classification for breast cancer being used in Italy has renamed DCIS to “DIN, Ductal Intraepithelial Neoplasia”.^61–63 Proponents for this approach note that the term “intraepithelial neoplasia” would be consistent with terminology currently used for precursor lesions in other organs such as the cervix, vulva, prostate and pancreas. It has also been argued that a name change to “DIN” would improve inter-observer agreement in the diagnosis of pre-invasive breast lesions and would eliminate the need to make the subjective distinction between atypical ductal hyperplasia and low grade DCIS. At this time, however, there are no data to indicate that changing the nomenclature would improve observer reproducibility.⁶⁰

It has also been suggested that removing “carcinoma” from the terminology for a disease which should not be able to spread to other parts of the body and threaten a woman’s life could lead to decreased anxiety among patients, improved risk perceptions and help in decision-making.⁸ However, there are no data at this point to suggest that a name change will have an effect on risk perceptions, anxiety/distress or decision-making.

Concern has been raised that the heterogeneity of DCIS, the use of treatment options that are similar to those used for patients with invasive breast cancer, and the limited ability to stratify risk using available clinical and pathologic parameters would limit the potential for a name change to improve risk perceptions and reduce anxiety. In addition, a change in terminology would not result in a change in the treatment options for patients with this disease and might lead to more confusion rather than less for patients and providers, particularly those providers at the periphery of the care of patients with breast cancer. The traditional terminology is well-established and deeply embedded with an associated extensive scientific literature. There are few citations from a limited number of authors considering the proposed DIN terminology. Some patient advocates have also expressed concern that changing the name could be construed by women as duplicitous and patronizing, and patients may ultimately experience similar distress to the term “neoplasia” in the DIN terminology compared with the term “carcinoma” in DCIS. It has been noted that lobular carcinoma in situ (LCIS) does not appear to generate the same anxiety or concerns as DCIS, possibly because treatment recommendations for women with DCIS are much more aggressive, a fact that would not change with a name change. Some advocates have suggested that research on biology, such as predictive and prognostic markers for DCIS that might help guide treatment decisions, and research to improve communication is a better use of resources than implementing a new nomenclature.

However, proponents suggest that a name change could be feasible if it were done in a phased approach: 1) first, discussions regarding terminology introduced in interdisciplinary settings; 2) next, pathology reports transitioning to include the traditional terminology along with DIN equivalent in parentheses; 3) subsequently, DIN designation being placed first on the pathology report followed by the traditional DCIS terminology in parentheses; and 4) finally, DIN terminology only being used in pathology reports.⁶⁰

In 2011, the World Health Organization Working Group for classification of tumors of the breast noted that the DIN terminology has not gained widespread acceptance, in part because no new diagnostic criteria are used, and suggested that a change in terminology would therefore not help with improving inter-observer reproducibility.¹⁵ In light of ongoing and future work in this area, the Working Group recommended that “classification of intraductal proliferative lesions should be viewed as an evolving concept that may be modified as additional molecular and genetic data become available.”¹⁵

In summary, while a change in terminology may be worthy of consideration in the future, there are no data to support the contention that a name change at the present time will reduce observer variability in diagnosis, reduce patient anxiety or assist patients and clinicians in choosing among the various treatment options for DCIS which will be the same regardless of the terminology used. Further, a name change should not be viewed as a substitute for communicating what DCIS means in terms of prognosis and treatment options. Many believe that clinical utility and patient benefit should drive the efforts for changing DCIS nomenclature and that at this time, efforts should be focused on ensuring that pathologists provide as accurate and consistent reporting of DCIS cases as possible.

Conclusions and Future Directions

The accuracy of perceived risk, anxiety, and decision-making among women with DCIS would likely be improved by better physician-patient communications about the disease, enhanced provision of psychosocial support, and better recognition and treatment of coexisting anxiety.^{43, 44} However, until we have a better understanding of the disease, predictors of risk and biologic behavior, and are able to develop more tailored therapy for individuals, the high level of uncertainty about the disease will continue to pose substantial challenges to informed decision-making and psychosocial outcomes.⁵¹ Nevertheless, to improve the situation, decisions aids are important tools to facilitate ongoing patient-clinician communication (not replace it) and further research among women with DCIS is clearly warranted. Careful attention to shared decision-making, eliciting and considering a woman’s preferences when helping her to make treatment decisions, as well as screening for and addressing anxiety in such patients, can improve care of individual patients.

Continued increases in DCIS diagnoses due to imaging advances and an aging population mandate improved communication about DCIS among professionals involved in the diagnosis and treatment of patients with the disease. Research is needed to understand better the biology and mechanisms of the progression of DCIS to invasive breast cancer and the factors which predict those subtypes of DCIS which do not progress, and to improve communications between patients and providers. There was no consensus among attendees at the workshop to support changing the nomenclature of DCIS. In the future, we seek to (1) evaluate the process by which other diseases made nomenclature changes and to determine the extent to which communication influenced implementation and quality of life of the patients; and, (2) obtain information from other countries including Italy where recent nomenclature changes have been adopted on the resulting effects of the changes on risk perceptions, psychosocial outcomes and decision-making.

Figure 2.

Patient Centered Communication Functions (from Epstein RM and Street RL, Jr.⁵⁰)

APPENDIX A

List of Workshop Participants (in alphabetical order)

Terri Ades, DNP, FNP-BC, AOCN, Director, Cancer Information, American Cancer Society, Atlanta, GA

D. Craig Allred, MD, Professor, Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, MO

Kimberly Andrews, Research Associate, American Cancer Society, Atlanta, GA

NeerajArora, PhD, Health Systems Analyst, Outcomes Research Branch, ARP, DCCPS, National Cancer Institute, Bethesda, MD

Otis W. Brawley, M.D., Chief Medical Officer, American Cancer Society, Atlanta, GA

Kara Smigel Croker, Communications Manager, Division of Cancer Prevention, National Cancer Institute, Bethesda, MD

Stephen B. Edge, MD, Alfiero Foundation Endowed Chair in Breast Oncology, Professor of Surgery and Oncology, Roswell Park Cancer Institute, Buffalo, NY

Joann G. Elmore MD, MPH, Professor of Medicine, Adjunct Professor Epidemiology, Univ. of Washington School of Medicine, Section Head of General Medicine, Harbor view Medical Center, Seattle, WA

Ted Gansler, MD, MBA, Medical Content Director, Editor, CA Cancer J Clin, American Cancer Society, Atlanta, GA

Len Lichtenfeld, MD, MACP, Deputy Chief Medical Officer, American Cancer Society, Atlanta, GA

WortaMcCaskill-Stevens, MD, MS, Program Director, Division of Cancer Prevention, National Cancer Institue, National Institutes of Health, Bethesda, MD

Sandra Millon Underwood, RN, PhD, University of Wisconsin Milwaukee, College of Nursing, Milwaukee, WI

Ann H. Partridge, MD, MPH, Associate Professor, Harvard Medical School, Dana-Farber Cancer Institute, Boston, MA

Barbara D. Powe. PhD, RN, Director, Communication Science, American Cancer Society, Atlanta, GA

Ada Patricia Romilly MD, Medical Director of Breast Imaging, Taylor Breast Health Center, Jackson Memorial Hospital, Miami, Florida

Debbie Saslow, PhD, Director of Breast and Gynecologic Cancer, American Cancer Society, Atlanta, GA

Stuart J. Schnitt, MD, Beth Israel Deaconess Medical Center, Boston, MA

Karen Sepucha, PhD, Health Decision Research Unit, MGH, Harvard Medical School, Massachusetts General Hospital, Boston, MA

Mary Lou Smith, Co-Founder, Research Advocacy Network, Plano, TX

Fattaneh A. Tavassoli, MD, Professor of Pathology/Obstetrics & Gynecology, Director, Women's Health Program/Gyn and Breast Pathology, Yale University School of Medicine, Dept. of Pathology, New Haven, CT

TheaTlsty, PhD, Professor of Pathology, University of California, San Francisco, San Francisco, CA

Umberto Veronesi, MD, Scientific Director, European Instittue of Oncology, Milano

Diana Zuckerman, PhD, President, National Research Center for Women & Families, Cancer Prevention and Treatment Fund, Washington, DC