Disparity in perceptions of disease characteristics, treatment effectiveness, and factors influencing treatment adherence between physicians and patients with myelodysplastic syndromes

Authors


  • Presented as abstract 724 at the American Society of Hematology 55th Annual Meeting; December 7-10, 2013; New Orleans, Louisiana.

Abstract

BACKGROUND

Previous studies have suggested that many patients with myelodysplastic syndromes (MDS) have an incomplete understanding of their disease, which may influence adherence to prescribed regimens and outcomes.

METHODS

To better understand physician and patient perceptions about MDS and MDS therapy, the authors conducted 2 surveys in February 2012: 1 for patients with MDS and 1 for health care professionals (HCPs) who cared for patients with MDS. Patient and HCP surveys consisted of 57 and 49 questions, respectively, assessing understanding of MDS, perceptions of specific treatments, barriers to treatment adherence, and treatment experience.

RESULTS

In total, 477 complete patient responses and 120 complete HCP responses were received. Among patient responders, 63% were aged ≥60 years, and 42% had received at least 1 disease-modifying therapy. Of the 61 physician responders, 57% practiced in an academic setting, and 43% practiced in the community; 71% of the 59 nonphysician HCPs worked in the community setting. Only 10% of patients agreed that MDS represented “cancer” compared with 59% of physicians and 46% of nonphysician HCPs (P < .001). Only 29% of patients reported that MDS was ever “curable” compared with 52% of physicians (P < .001). Physicians viewed the potential benefits of active therapy as greater than patients, but patients perceived the actual treatment experience more positively than physicians and differed from physicians in perceived reasons for stopping therapy.

CONCLUSIONS

Physicians, nonphysician HCPs, and patients with MDS have disparate views of MDS characteristics and the value and limitations of treatments for MDS. Improved communication and education may increase understanding and achieve better treatment adherence and patient outcomes. Cancer 2014;120:1670–1676. © 2014 American Cancer Society.

INTRODUCTION

The myelodysplastic syndromes (MDS) include a diverse group of hematologic neoplasms collectively characterized by bone marrow failure and peripheral blood cytopenias, oligoclonal hematopoiesis, characteristic cellular dysmorphology, and genomic instability with a tendency to progress to acute myeloid leukemia (AML).[1] These chronic diseases, which have a heterogeneous natural history but an overall median survival of about 3 years, place a great symptomatic and psychosocial burden on patients and an economic burden on health care systems.[2, 3] Although past registry-based reports using the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) database have suggested an incidence of 10,000 to 12,000 MDS cases per year in the United States, more recent estimates using Medicare claims data suggest that MDS are diagnosed in at least 40,000 to 50,000 patients per year, which would place MDS among the 10 most frequently encountered adult neoplasms.[4, 5]

Most patients with MDS—even those with lower risk disease features at the time of diagnosis—will die of infectious or hemorrhagic complications related to their disease.[6, 7] The only potentially curative therapy, allogeneic hematopoietic stem cell transplantation (HSCT), is currently attempted in less than 5% of patients because of the advanced age and comorbid conditions present in most patients and perceptions about the danger of HSCT.[8, 9] There are currently 3 drugs approved for MDS by the US Food and Drug Administration (FDA) that offer symptomatic benefit and slow disease progression in some patients: azacitidine (AZA), decitabine (DAC), and lenalidomide (LEN). In addition, the erythropoiesis-stimulating agents epoetin alpha and darbepoetin alpha are widely used off-label in MDS to augment erythrocyte production.

Previous studies have suggested that many patients with MDS, as well as their health care providers, have an incomplete understanding of fundamental concepts about these diseases and potential treatment options.[10-12] In addition, contemporary drug therapies for MDS require repeated treatment administration cycles in specific schedules to achieve clinical effect and to durably maintain hematopoietic improvement, and incomplete adherence or premature discontinuation of therapy may result in lack of benefit for patients who might have responded to prescribed therapy.[13, 14] To better understand physician and patient perceptions about MDS and how these influence decisions about MDS therapy, we conducted 2 surveys in February 2012: 1 for patients with MDS and 1 for health care providers (HCPs) registered with the nonprofit Aplastic Anemia and Myelodysplastic Syndrome International Foundation (AA&MDS IF) (Rockville, Md).

MATERIALS AND METHODS

The study protocol was approved by a central institutional review board. The survey was sent by an embedded webpage link to patients and HCPs with e-mail addresses on file with the AA&MDS IF. A $50 Amazon.com gift card was offered to the first 200 respondents in each group. Online consent to participate in the survey constituted informed consent.

Patient and HCP surveys consisted of 57 and 49 questions, respectively, assessing understanding of MDS, perceptions of specific treatments, perceived barriers to treatment adherence, and overall treatment experience. The surveys were developed in consultation with MDS physicians, patients, and patient advocacy group members and were piloted in a small focus group of patients and HCPs whose answers are not included in these results, so that ambiguous or confusing questions could be removed or modified. Pilot testing of the surveys demonstrated that they could be completed in 20 to 30 minutes.

Previous studies have suggested that patient self-reporting of International Prognostic Scoring System (IPSS) risk score, disease subtype, and karyotype are limited because >40% of patients are unaware of those parameters; therefore, IPSS, classification, and cytogenetic data were not requested in this survey.[3, 10] For qualitative questions in which responses were in the form of a 6-point Likert scale ranging from “strongly disagree” to “strongly agree,” the 3 answers nearest “strongly agree” were considered together as assent to the statement. Data were analyzed using proportions, means, and medians; and comparisons between groups were calculated using the Pearson chi-square test where appropriate. A 2-sided P value < .05 was used to determine significance.

RESULTS

Altogether, 477 complete patient responses were received from participants living in 42 US states, and 120 complete HCP responses were received from 24 US states. Survey invitations were sent to 4039 patients and 4594 HCPs, so the survey response rates for patients and HCPs were 11.8% and 2.6%, respectively, not accounting for individuals who had died or for erroneous e-mail addresses.

Of the patient responders, 247 (52%) were men; 63% were aged ≥60 years, and 28% were aged ≥70 years; 32% described themselves as members of minority groups (26% Hispanic or Latino, 6% African American or other); and the median time from MDS diagnosis was 5 years (range, 0-32 years). These are typical demographics for MDS as a whole, except for a slightly longer median time from diagnosis and younger age (64 years vs 71 years[8]) and a higher proportion of minority patients. Only 12% of patients stated that they had completed the survey with assistance, mostly from a spouse (56% of assisted surveys); and 88% completed the survey on their own.

Of the 61 physician responders, 35 (57%) practiced in an academic setting in which clinical trials were conducted, 26 (43%) practiced in the community setting, and 48% reported that they acquired between 5 and 19 new MDS patients per year. Because survey responses from academic and community physicians did not differ significantly, the results were combined. Nonphysician HCP survey responders included 46 nurses or nurse practitioners and 13 physician assistants or social workers. Among nonphysician HCPs, 71% worked in the community setting, and 39% reported that they acquired between 5 and 19 new MDS patients per year, whereas the rest acquired <5 new MDS patients per year. Because subgroups of nonphysician HCPs were small and did not differ significantly, the nonphysician HCP results were combined.

Only 10% of patients agreed that MDS represented “cancer” compared with 46% of nonphysician HCPs and 59% of physicians, who answered that they described MDS as cancer to their patients (P < .001 for both physician vs patient comparison and nonphysician HCP vs patient comparison). The majority of patients reported instead that their MDS was described to them as a bone marrow failure syndrome (60%), a blood disease (38%), a form of anemia (23%), or a preleukemic condition (17%; these answers summed to >100% because patients were permitted multiple responses to this question). Only 29% of patients reported that they believed MDS was ever a “curable” condition compared with 33% of nonphysician HCPs and 52% of physicians (P < .001 for patient vs physician comparison, P = .025 for physician vs nonphysician HCP comparison, P value nonsignificant [NS] for nonphysician HCP vs patient comparison). Physician responses about the number of treatment cycles that they believed comprised an adequate therapeutic trial were inconsistent with National Comprehensive Cancer Network (NCCN) guidelines[15] for AZA, DAC, and LEN in 35%, 24%, and 13% of surveys, respectively.

Although 56% of patients had received transfusions, hematopoietic growth factors, or other supportive care modalities (eg, iron chelation, androgens), 42% of patients (ie, 200 of the 477 respondents) had received at least 1 potentially disease-modifying therapy. Twenty-four percent of survey respondents had received AZA (n = 115), 11% of patients had received DAC (n = 54); 14% of patients had received LEN (n = 69), and 11% of patients underwent HSCT (n = 51; this total is >42% because some patients received multiple treatments). Compared with published series, more patients in this survey had undergone HSCT.[8, 16] This difference could be because survey respondents may have had a higher performance status and may have been younger than the average patient with MDS and, thus, were more likely to be HSCT candidates, or because HSCT recipients may have been more invested in their care because of the nature of transplantation and, thus, were more likely to complete the survey.

The median number of AZA cycles delivered to patients was 6 (113 of 115 patients reported these data), whereas 41% of patients had received <6 cycles of AZA. The median number of DAC cycles administered (all 54 patients who received DAC reported treatment length data) was 6 cycles, and 33% received <4 cycles of DAC. The median number of LEN cycles administered (with 68 of 69 patients reporting) was 6 cycles, and 19% received <2 monthly treatment cycles of LEN.

Physician, nonphysician HCP, and patient perceptions of specific MDS therapies were significantly different, especially regarding health-related quality of life during treatment, adverse events, and the impact of treatment on patient activities (Table 1). Physicians and, to a lesser extent, nonphysician HCPs viewed the potential benefits of active treatment as being significantly greater than did patients; however, patients perceived the actual treatment experience more positively than physicians or nonphysician HCPs (Table 2). Nonphysician HCPs were less sanguine about the benefit of specific therapies and were more aware of the burdens on patients than physicians, possibly because of more frequent contact with patients undergoing therapy (Table 1).

Table 1. Physician, Nonphysician Health Care Professional, and Patient Perceptions of Specific Therapies for Myelodysplastic Syndromes
 Azacitidine
 Respondents, % Respondents, % Respondents, % 
Statements About Drug to Which Respondents AssentedPhysicians, n = 55Patients, n = 115P for ComparisonPhysicians, n = 55Nonphysicians, n = 51P for ComparisonNonphysicians, n = 51Patients, n = 115P for Comparison
This treatment can improve patients' quality of life8963.0018980NS8063.030
This treatment makes patients (makes me) feel better while taking it6229< .0016253NS5329.003
This treatment has no side effects724.008724.0192424NS
This treatment makes regular activities difficult on treatment days2745.0252741NS4145NS
This treatment makes regular activities difficult for days following treatment administration2543.0302537NS3743NS
 Decitabine
 Respondents, % Respondents, % Respondents, % 
Statements About Drug to Which Respondents AssentedPhysicians, n = 39Patients, n = 54P for ComparisonPhysicians, n = 39Nonphysicians, n = 43P for ComparisonNonphysicians, n = 43Patients, n = 54P for Comparison
This treatment can improve patients' quality of life8565.0348567NS6765NS
This treatment makes patients (makes me) feel better while taking it6246NS6256NS5646NS
This treatment has no side effects528.0055146NS1428NS
This treatment makes regular activities difficult on treatment days2641NS2630NS3041NS
This treatment makes regular activities difficult for days following treatment administration3144NS3133NS3344NS
 Lenalidomide
 Respondents, % Respondents, % Respondents, % 
Statements About Drug to Which Respondents AssentedPhysicians, n = 40Patients, n = 69P for ComparisonPhysicians, n = 40Nonphysicians, n = 40P for ComparisonNonphysicians, n = 40Patients, n = 69P for Comparison
  1. Abbreviations: NS, nonsignificant.

This treatment can improve patients' quality of life8057.0138080NS8057.013
This treatment makes patients (makes me) feel better while taking it6335.0056350NS5035NS
This treatment has no side effects525.009515NS1525NS
This treatment makes regular activities difficult on treatment days3026NS3025NS2526NS
This treatment makes regular activities difficult for days following treatment administration6022< .0016033.0143322NS
Table 2. Physician, Nonphysician Health Care Professional, and Patient General Perceptions of Myelodysplastic Syndrome Therapy
 Respondents Assenting, %  
Clinician StatementPhysicians, n = 61Patients, n = 200P for ComparisonCorresponding Patient Statement
My patients' lives are better because I prescribed treatment9579.004My life is better because I received treatment
My patients are glad they have received treatment9382.030I'm glad I had the treatment
Treatment cured my patients' MDS4425.003Treatment cured my MDS
Treatment is uneventful1634.010Treatment was uneventful
Getting treatment is easy2151< .001Getting treatment was easy
 Respondents Assenting, %  
Clinician StatementNonphysicians, n = 59Patients, n = 200P for ComparisonCorresponding Patient Statement
My patients' lives are better because I prescribed treatment7879NSMy life is better because I received treatment
My patients are glad they have received treatment8382NSI'm glad I had the treatment
Treatment cured my patients' MDS3225NSTreatment cured my MDS
Treatment is uneventful1934.029Treatment was uneventful
Getting treatment is easy2951.003Getting treatment was easy
 Respondents Assenting, %  
Clinician StatementPhysicians, n = 61Nonphysicians, n = 59P for Comparison
  1. Abbreviations: MDS, myelodysplastic syndrome; NS, nonsignificant.

My patients' lives are better because I prescribed treatment9578.006
My patients are glad they have received treatment9383NS
Treatment cured my patients' MDS4432NS
Treatment is uneventful1619NS
Getting treatment is easy2129NS

Sixty-nine percent of physicians reported having recommended stopping treatment before the completion of a planned treatment course, most commonly because of adverse events and because the burden of therapy outweighed the benefit. Physicians and nonphysician HCPs were approximately twice as likely as patients to report burden of treatment or patient symptoms as a reason for therapy cessation (P < .004 across reasons) (Table 3). The physicians were not asked details about planned courses of therapy.

Table 3. Physician, Nonphysician Health Care Professional, and Patient Perceptions About Reasons for Discontinuing Most Recent Therapy
 Respondents Assenting, % 
Clinician Statements (Patient Comparative Statement)Physicians, n = 42Patients, n = 116P for Comparison
Burden of treatment outweighed benefit to patient (me)6935< .001
Burden of treatment was too great on the family/caregiver5023.001
Patient fatigue was too great to continue6033.002
Treatment made the patient (me) feel too sick to continue7933< .001
Treatment side effects interfered with the patient's (my) regular activities7435< .001
 Respondents Assenting, % 
Clinician Statements (Patient Comparative Statement)Physicians, n = 42Nonphysicians, n = 27P for Comparison
Burden of treatment outweighed benefit to patient (me)6978NS
Burden of treatment was too great on the family/caregiver5033NS
Patient fatigue was too great to continue6063NS
Treatment made the patient (me) feel too sick to continue7978NS
Treatment side effects interfered with the patient's (my) regular activities7452NS
 Respondents Assenting, % 
Clinician Statements (Patient Comparative Statement)Nonphysicians, n = 27Patients, n = 116P for Comparison
  1. Abbreviations: NS, nonsignificant.

Burden of treatment outweighed benefit to patient (me)7835< .001
Burden of treatment was too great on the family/caregiver3323NS
Patient fatigue was too great to continue6333.004
Treatment made the patient (me) feel too sick to continue7833< .001
Treatment side effects interfered with the patient's (my) regular activities5235NS

Most patients (81%) reported that it was the recommendations of their physician that had the greatest influence on their treatment decisions. Reported reasons for discontinuation of therapy diverged between patients and physicians, including burden of treatment exceeding the benefit to the patient as well as perceptions that the impact on the patient and family was too great.

DISCUSSION

Health-related quality of life and clinical outcomes for patients with MDS are poor. The median survival in patients diagnosed with higher risk forms of the disease is <2 years, and MDS-associated complications are the leading cause of death even among lower risk patients.[6, 17] Whereas an incomplete pathobiologic understanding of MDS by the biomedical community and the paucity of available highly effective MDS treatments are the major factors contributing to these poor outcomes, suboptimal use of currently available therapies, incomplete understanding of disease by patients and HCPs, and premature discontinuation of potentially effective treatments may also be important factors influencing MDS-associated outcomes.

Because all 3 FDA-approved therapies for MDS require a period of time to have a favorable effect, early discontinuation of therapy may lower response rates.[13] Current NCCN guidelines suggest that an adequate therapeutic trial of LEN is 8 to 12 weeks (the median time to response with this agent is 5 weeks); whereas it is appropriate to treat with 6 monthly cycles of AZA and with 4 monthly cycles of DAC (if tolerated), at which time 90% of patients who will respond to AZA or DAC will have begun to exhibit a favorable response.[15, 18-20]

However, available data suggest that discontinuation of therapy before these time points is common, and physicians in this survey frequently reported that what they considered an “adequate trial” was less than NCCN guideline recommendations (ie, <6 cycles for AZA and <4 cycles for DAC). In the AVIDA registry, for example, which captured data on 380 patients who received AZA between 2006 and 2008 (ie, 2-4 years after FDA drug approval), only 17% of patients received the FDA-approved 7-day administration cycle, and the median number of treatment cycles administered was 4.[21] In 2012, only approximately 66% of patients who were prescribed LEN refilled their first prescription (ie, continued therapy for at least 4 weeks), and only 50% refilled their second LEN prescription (Celgene, data on file). Although some of the early discontinuations are appropriate and are caused by progressive disease or unresolvable intolerance of therapy, other discontinuations may be a result of inadequate education or of HCPs and patients giving up too soon.

These survey results suggest that HCPs may recommend discontinuation of therapy because of perceptions that patients are doing more poorly on therapy than they actually are, whereas some patients may remain willing to continue therapy if adverse events are addressed and if there is still a chance of response. Consistent with this finding, most patients report that a physician's recommendation was the primary factor in therapy discontinuation. Factors contributing to MDS therapy discontinuation have not previously been assessed outside the context of prospective clinical trials of specific therapies in carefully screened and selected patients.

Whether MDS represents “cancer” is a somewhat controversial topic, although the World Health Organization classifies MDS as neoplastic, because the biologic characteristics, including clonality, resemble other neoplasms.[22, 23] This semantic issue persists because MDS can remain stable for several years and because immune-mediated suppression of hematopoiesis is an important contributor to MDS pathophysiology in some cases.[22, 23] The definition of MDS as neoplastic versus non-neoplastic has practical consequences, for example, with respect to MDS patients' eligibility to make claims against cancer-specific indemnity policies, investigators' eligibility for funding by organizations dedicated to cancer research, and the accuracy of registry data. Physician and nonphysician HCP survey respondents may have disbelieved the World Health Organization classification or may have believed that using the word “cancer” would have been harmful to patients; alternatively, patients may have had MDS described to them as cancer but chose not to remember it, either because of fear of cancer or because they were aware of ways in which MDS is distinct from other cancers.

Only 29% of patients believed that MDS could ever be “cured,” although the median age of respondents was 64 years, well within the range in which reduced-intensity-conditioning allogeneic HSCT is still routinely practiced. This suggests that patients with MDS are incompletely educated about HSCT. Some patients who could have been candidates for HSCT may never have been offered this therapy, and underuse of HSCT among appropriate candidates may contribute to poor outcomes.[9] Because nonphysician HCP responses to this question were more similar to patient responses than to physician responses, and because nonphysician HCPs have a critical role in patient education, MDS-focused educational initiatives directed specifically to nonphysician HCPs also should be considered.

Limitations of this study include the online nature of the survey, which restricted participation to English-speaking patients with internet access who had previously had contact with the AA&MDS IF and who were healthy enough to be able to take the survey. In addition, patients and HCPs may have interpreted questions in various ways, with HCPs answering based on aggregate experience (and potentially influenced by bias from more recent experience), whereas patients may gave answered on the basis of their own individual case. Because patients' medical records were not reviewed, survey responses could not be correlated with actual treatment administration data or with clinical outcomes, and HCP responses could not be paired directly with responses from their own patients. Likewise, the relatively small cohort of HCPs precluded more comprehensive statistical analyses relating to location and type of practice, impact of practice volume, and years of experience. These data also are unlikely to be relevant outside the United States, because treatment availability and care patterns differ by region.

Furthermore, only a small percentage of solicited patients and HCPs responded to these surveys; therefore, the current results may not be generalizable to the entire population of MDS patients and HCPs. Still, the participation rates for both are consistent with other, similar studies[14]; and, because the patients and HCPs who respond to these surveys are likely to be more engaged in MDS treatment than the general population, the broader populations of patients with MDS and HCPs are likely to have even less of an understanding of the disease and perhaps a higher likelihood of prematurely discontinuing therapy.

Despite these limitations, this survey demonstrates that physicians, nonphysician HCPs, and patients with MDS have disparate views of MDS characteristics and of the value and limitations of treatment for MDS. Improved communication and education may increase understanding of disease and achieve better results, including improved treatment outcomes.

FUNDING SUPPORT

This study was funded in part by an unrestricted educational grant to the Aplastic Anemia & Myelodysplastic Syndromes International Foundation from Celgene Corporation and by the Edward P. Evans Foundation. The authors designed and conducted the study and wrote the article without influence from the sponsor.

CONFLICT OF INTEREST DISCLOSURES

Dr. Steensma reports personal fees for service on scientific advisory boards from Astex, Boehringer Ingelheim, Novartis, Incyte, Genoptix, and Opsona; personal fees for service on a Data Safety Monitoring Committee from Amgen; personal fees for service on a Registry Study Steering Committee from Celgene; personal fees for editorial services from the American Society of Hematology and the American Society of Clinical Oncology; personal fees from the New England Journal of Medicine for commentary preparation; personal fees for consulting from Gerson Lehman Group; personal fees for continuing medical education presentation from Knowledge Oncology, LLC; nonremunerated service on the MDS Foundation Board of Directors and on the Aplastic Anemia & Myelodysplastic Syndrome International Foundation Medical Advisory Board; and grants from the Edward P. Evans Foundation; he has stock in Ariad; and he is an employee of the Dana-Farber Cancer Institute and Harvard University. Dr. Komrokji reports clinical research funding and personal fees from Celgene and personal fees from Novartis. Dr. Sekeres reports other compensation from Celgene and Amgen.

Ancillary