ilMimopf

0©d

Patient-ReportedOutcomes in Supportive Care

Emma Bateman and Dorothy Keefe

Traditionally, anticancer therapy has focused on eradication of neoplastic tissue, predominantly byinvasive and/or toxic treatments. In modern studies, patient-reported outcomes (PROs) havebecome more common, and give a true picture of toxicity. Increased consideration of subjectivepatient perspectives of anticancer treatments has allowed a notable shift within supportive oncol-ogy. Disparity exists between physician and patient perspectives of symptom severity, despiteseveral common scoring methods. PROs are vital tools in the overall assessment of chronicillnesses, including cancer and associated treatments. Synergistic assessments of objective andsubjective observations of symptoms and function are most accurate. PROs include informationcollected either in a clinic or by a diary system. Patient self-reporting, like any other assessment ofhealth status, is not an absolute measure. Electronic data collection is an increasingly useful way tomonitor PROs. Factors that influence quality of life (QOL) are predominantly subjective experi-ences, and can occur concurrently with pre-existing symptoms, which increases symptom burden.There are several validated systems for assessing PROs; some are concerned with specific condi-tions like mucositis (Oral Mucositis Weekly Questionnaire [OMDQ]), whereas others cover chronicillness in general (Patient-Reported Outcome Measurement Information System [PROMIS]). ThePROMIS framework was developed by the National Health Institute (NHI) to standardize self-reported health measurements within chronic illnesses, including pain, fatigue and emotionaldistress. The general Assessment of Cancer Therapy (FACT-G) scale was developed to assess manydifferent types of cancer; we will discuss use in oral mucositis as a model. There is more tomeasuring toxicity than the clinician’s objective view of the patient experience. There is still muchto be done to validate all the necessary PRO tools so that we can competently measure both toxicityand toxicity-reduction strategies. Current systems to assess PROs continue to have a very positiveimpact on supportive oncology.Semin Oncol 38:358-361 © 2011 Elsevier Inc. All rights reserved.

cttpsimPipy

oo(1m

Traditionally, anticancer therapy has focused oneradication of neoplastic tissue, predominantlyby invasive and/or toxic treatments. The toxic-

ties that patients needed to tolerate to achieve theonged-for cure were not always addressed in detail.

ore recently, the focus of treatment has broadened tonclude the side effects of cancer and cancer treat-

ents; such supportive oncology is now a major focusf healthcare. When toxicity was first considered, theatient role in determining severity was limited; it was

elt that “objective” measures must be used and that

Mucositis Research Group, Discipline of Medicine, Faculty of HealthSciences, University of Adelaide, Adelaide, Australia.

Statement of conflict of interest: the authors have no conflicts of interest.Address correspondence to Professor Dorothy Keefe, Level 4, East

Wing, Royal Adelaide Hospital, North Terrace, Adelaide SA 5000,Australia. E-mail: Dorothy.Keefe@health.sa.gov.au.

270-9295/ - see front matter2011 Elsevier Inc. All rights reserved.

oi:10.1053/j.seminoncol.2011.03.003

Se358

linicians were more objective. It has been realizedhat patients are good at determining severity of symp-oms. While patient-reported symptoms scores oftenarallel the severity measured by clinicians, patientsuffer more than clinicians appreciate. In modern stud-es, patient reported outcomes (PROs) have become

ore common, and give a true picture of toxicity.atient-reported toxicity also tracks reduction in qual-

ty of life (QOL). The expectation is that reduction inatient-reported toxicity would improve QOL. This haset to be conclusively proven.

A focus on the balance between quality and quantityf life associated with cancer treatments has devel-ped. With funding from the National Cancer InstituteNCI) of clinical trials cancer cooperative groups in955, the objectives of cancer clinical trials were pri-arily to increase survival and decrease morbidity.1 It

would be another three decades before measures ofQOL were considered in the context of efficacy of

anticancer treatments. In 1988, the NCI Cancer Ther-

minars in Oncology, Vol 38, No 3, June 2011, pp 358-361

apts

cmfd

aipohCaarsc

oetepiv

bsdPle

jpdgvt

Patient-reported outcomes in supportive care 359

apy Evaluation Program (CTEP) modified their missionstatement, declaring, “research aimed at improving sur-vival and QOL for persons with cancer is of the highestpriority.”2

Increased consideration of subjective patient per-spectives of anticancer treatments has allowed a nota-ble shift within supportive oncology. There is oftenincongruity between observable, clinical manifesta-tions and patient perception of pain, nausea, vomitingand diarrhea, mucositis, and fatigue. All of them arehallmarks of antineoplastic therapies, and all affectQOL.3–5 Many “objective” evaluations by physiciansnd nurses underestimate pain and dysfunction.5,6 Dis-arity exists between physician and patient perspec-ives of symptom severity, despite several commoncoring methods.7

PATIENT-REPORTED OUTCOMES

PROs are vital tools in the overall assessment ofchronic illnesses, including cancer and associated treat-ments.8,9 These PROs are defined as any aspect of anindividual’s health status that comes directly from thepatient, without influence, interpretation, or modifica-tion by any other observer.10 It has been shown thatlinician-reported outcomes (CROs) of anticancer treat-ents may improve prediction of risk for adverse ef-

ects longitudinally, PROs are much better indicators ofaily patient status.11 One problem associated with

many side effects of cancer treatments is that symp-toms like nausea, vomiting, and fatigue occur awayfrom the clinical setting.12 In evidence-based, objectivessessment of symptoms, this can lead to under-report-ng of symptoms. Consequently, up-to-date patient re-orting is invaluable to detect the severity and coursef adverse symptoms, and to allow timely treatment. Itas become evident that there are limitations of bothROs and PROs. Synergistic assessments of objectivend subjective observations of symptoms and functionre most accurate. They define the exact patient status,ather than traditional, evidence-based objective mea-urements used alone. This allows for better supportiveare.

PROs include information collected either in a clinicr by a diary system. These may include symptom andvent logs, single-item outcome measures, instrumentso measure health-related QOL, health status, adher-nce, and overall satisfaction with treatment.13 In someatient-reporting systems, caregivers contribute their

ndependent observations of symptoms like nausea andomiting.3 Observations by nursing staff are useful also,

as they are more likely to have frequent patient con-tact. People often are more comfortable discussingadverse effects with a nurse rather than a clinician.6

Patient self-reporting, like any other assessment ofhealth status, is not an absolute measure. The advan-

tage of subjective perceptions for patient status mea-

sures may be counteracted by this subjectivity. Forexample, perception of very high or low pain levelswill be interpreted and reported differently by differentpatients.14 Compliance dependent on patient experi-ence is another important issue; for example, the verysick may be unable to report their status.15 Anotherconcern in assessing PROs is the burden of these mea-surements on the patient and research personnel.14 A

alance between time requirements and precision en-ures that additional stress is minimized. Electronicata collection is an increasingly useful way to monitorROs. Access to these systems, as well as computer

iteracy (particularly in the elderly), may influence theirffectiveness.

Factors that influence QOL are predominantly sub-ective experiences, and can occur concurrently withre-existing symptoms, which increases symptom bur-en.16 For example, copious quantities of oropharyn-eal mucous with oral mucositis can cause nausea,omiting, and fatigue independently of those symp-oms caused directly by the anticancer therapy.15,17

Such increases in symptom burden may exacerbatedeclines in physical and psychological QOL, furthercompounding the adverse effects of antineoplastictherapy. Evaluation of PROs, particularly if performedearly in a treatment regimen, may decrease symptomburden and increase QOL and recovery.

ASSESSMENT OFPATIENT-REPORTED OUTCOMES

As with any assessment module to determine effi-cacy versus toxicity of a treatment, there needs to beregulation of systems using PROs, particularly standard-ized training.1,13,18,19 There are several validated systemsfor assessing PROs; some are concerned with specificconditions like mucositis (OMDQ), whereas otherscover chronic illness in general (PROMIS).

Patient-Reported OutcomesMeasurement Information System(PROMIS; National Health Institute)

The PROMIS framework was developed by the Na-tional Health Institute (NHI) to standardize self-reportedhealth measurements within chronic illnesses, includingpain, fatigue, and emotional distress. Additional domainsaddressing cancer- and supportive oncology-specific as-sessments are being developed.20–22

Functional Assessmentof Cancer Therapy (FACT) System

The general FACT (FACT-G) scale23 was developedto assess many different types of cancer. The nomen-clature of each FACT system reflects each type of can-cer; for example, FACT-L designates that to determine

PROs in lung cancer.24 Each reflects symptoms and

boilNcC

Mbg

360 E. Bateman and D. Keefe

experiences that may be specific to particular cancertypes.

PRO SYSTEMS FOR MUCOSITIS

The painful and potentially life-threatening sequelaeof mucositis are dose-limiting effects of antineoplastictherapies Complications of mucositis often interrupttreatment modules and further compromise the effec-tiveness of these treatments. Thus far, mucositis man-agement is supportive; mucositis symptom control(such as pain, nausea, vomiting, diarrhea, and fatigue)may be improved by standardized PROs.

National Cancer InstituteCommon Terminology Criteriafor Adverse Effects (NCI CTCAE)

In the United States, the NCI CTCAE is the mostcommonly used assessor of mucositis.17 It is clinician-

ased, and reports clinical and functional observations;ne of the principal drawbacks is inter-assessor variabil-

ty.25 Adjunct modules tailored to mucositis symptoms,ike the M.D. Anderson Symptom Inventory–Head andeck (MDASI-HN) module, more accurately reflect mu-ositis time-course and severity changes than the NCITCAE alone.17

Oral Mucositis Daily Questionnaire (OMDQ)

A study on the effects of palifermin on mucositisused not only CROs, but also measurements of func-tionality and patient self-assessment (via OMDQ).26 Sim-ilar comparisons were made between clinicians, func-tionality assessments, and subjective observationsregarding mucositis grade. It was shown that patientsreported symptom onset, peak, and resolution 1 to 3days earlier than clinicians.26,27 A weekly version (Oral

ucositis Weekly Questionnaire [OMWQ]) also haseen evaluated in head and neck cancer, and showedood correlation between CROs and PROs.28 Similar

observations of delayed symptom reporting by clini-cans were observed, which may be attributed to symp-toms occurring out of the clinical setting. This furtherunderscores the importance of real-time PROs. Theadult-validated OMDQ also has been used in a pediatricsetting; an adaptation of this assessment evaluated mu-cositis symptoms in children.29

Patient-Reported OralMucositis Symptom (PROMS)

The PROMS scale was developed and validated.30

This study compared the PROMS scale with FACT andclinician-rated oral mucositis (Visual Analogue Scale—Oral Mucositis Assessment Scale) and showed a highinternal reliability and strong correlation with clinical

assessments.

SUMMARY

There is more to measuring toxicity than the clini-cian’s objective view of the patient experience. Goneare the days of ignoring toxicity in response to per-ceived powerlessness to prevent it. We now have afuller comprehension of the vast difference betweenobjective observation and subjective experience. Thisis particularly true regarding perception of pain, andallows more accurate evaluation of patient treatmentand overall well-being. There is still much to be done tovalidate all the necessary PRO tools so that we cancompetently measure both toxicity and toxicity-reduc-tion strategies. Future clinical trials of cancer treat-ments should all contain PROs for toxicity. Electroniccapture of real-time symptoms would be an excellentstep forward,31,32 and would enable clinicians to re-spond to patient symptoms in a more timely manner,hopefully (but not yet certainly) improving QOL.

Additional population studies to fine-tune PRO as-sessments would allow more accurate interpretation,considering variables that may affect patient reporting.These would include age, gender, socioeconomic sta-tus, and geographic distance from clinics. It is possiblethat these modifications may influence future PRO as-sessment modules. Current systems to assess PROs con-tinue to have a very positive impact on supportiveoncology.

REFERENCES1. Bruner DW, Bryan CJ, Aaronson N, et al. Issues and

challenges with integrating patient-reported outcomes inclinical trials supported by the National Cancer Institute-sponsored clinical trials networks. J Clin Oncol. 2007;25:5051–7.

2. Clinical Trials Cooperative Group Program–Cancer Ther-apy Evaluation Program: guidelines. Bethesda, MD: Na-tional Cancer Institute, Division of Cancer Treatment;1988.

3. Grunberg SM, Boutin N, Ireland A, et al. Impact of nau-sea/vomiting on quality of life as a visual analogue scale-derived utility score. Support Care Cancer. 1996;4:435–9.

4. Fromme EK, Eilers KM, Mori M, Hsieh YC, Beer TM. Howaccurate is clinician reporting of chemotherapy adverseeffects? A comparison with patient-reported symptomsfrom the Quality-of-Life Questionnaire C30. J Clin Oncol.2004;22:3485–90.

5. Vickers AJ, Salz T, Basch E, et al. Electronic patientself-assessment and management (SAM): a novel frame-work for cancer survivorship. BMC Med Inform DecisMak. 2010;10:34.

6. Cirillo M, Venturini M, Ciccarelli L, et al. Clinician versusnurse symptom reporting using the National Cancer In-stitute-Common Terminology Criteria for Adverse Eventsduring chemotherapy: results of a comparison based onpatient’s self-reported questionnaire. Ann Oncol. 2009;20:1929–35.

7. Elting LS, Keefe DM, Sonis ST, et al. Patient-reported

measurements of oral mucositis in head and neck cancer

Patient-reported outcomes in supportive care 361

patients treated with radiotherapy with or withoutchemotherapy: demonstration of increased frequency,severity, resistance to palliation, and impact on quality oflife. Cancer. 2008;113:2704–13.

8. Granda-Cameron C, Viola SR, Lynch MP, Polomano RC.Measuring patient-oriented outcomes in palliative care:functionality and quality of life. Clin J Oncol Nurs. 2008;12:65–77.

9. Wagner LI, Wenzel L, Shaw E, Cella D. Patient-reportedoutcomes in phase II cancer clinical trials: lessonslearned and future directions. J Clin Oncol. 2007;25:5058–62.

10. Lipscomb J, Reeve BB, Clauser SB, et al. Patient-reportedoutcomes assessment in cancer trials: taking stock, mov-ing forward. J Clin Oncol. 2007;25:5133–40.

11. Basch E, Jia X, Heller G, et al. Adverse symptom eventreporting by patients vs clinicians: relationships withclinical outcomes. J Natl Cancer Inst. 2009;101:1624–32.

12. Molassiotis A, Coventry PA, Stricker CT, et al. Validationand psychometric assessment of a short clinical scale tomeasure chemotherapy-induced nausea and vomiting:the MASCC antiemesis tool. J Pain Symptom Manage.2007;34:148–59.

13. Willke RJ, Burke LB, Erickson P. Measuring treatmentimpact: a review of patient-reported outcomes and otherefficacy endpoints in approved product labels. ControlClin Trials. 2004;25:535–52.

14. Garcia SF, Cella D, Clauser SB, et al. Standardizing pa-tient-reported outcomes assessment in cancer clinicaltrials: a patient-reported outcomes measurement infor-mation system initiative. J Clin Oncol. 2007;25:5106–12.

15. Patrick DL, Ferketich SL, Frame PS, et al. National Insti-tutes of Health State-of-the-Science Conference State-ment: Symptom Management in Cancer: Pain, Depres-sion, and Fatigue, July 15–17, 2002. J Natl Cancer Inst.2003;95:1110–7.

16. Cleeland CS. Symptom burden: multiple symptoms andtheir impact as patient-reported outcomes. J Natl CancerInst Monogr. 2007;37:16–21.

17. Rosenthal DI, Mendoza TR, Chambers MS, et al. TheM.D. Anderson symptom inventory-head and neck mod-ule, a patient-reported outcome instrument, accuratelypredicts the severity of radiation-induced mucositis. Int JRadiat Oncol Biol Phys. 2008;72:1355–61.

18. Morris LA, Miller DW. The regulation of patient-reportedoutcome claims: need for a flexible standard. ValueHealth. 2002;5:372–81.

19. Lasch KE, Marquis P, Vigneux M, et al. PRO develop-ment: rigorous qualitative research as the crucial foun-

dation. Qual Life Res. 2010;19:1087–96.

20. Gershon RC, Rothrock N, Hanrahan R, Bass M, Cella D.The use of PROMIS and assessment center to deliverpatient-reported outcome measures in clinical research.J Appl Meas. 2010;11:304–14.

21. Cella D, Yount S, Rothrock N, et al. The Patient-ReportedOutcomes Measurement Information System (PROMIS):progress of an NIH Roadmap cooperative group duringits first two years. Med Care. 2007;45(Suppl 1):S3–11.

22. Cella D, Riley W, Stone A, et al. The Patient-ReportedOutcomes Measurement Information System (PROMIS)developed and tested its first wave of adult self-reportedhealth outcome item banks: 2005–2008. J Clin Epide-miol. 2010;63:1179–94.

23. Cella DF, Tulsky DS, Gray G, et al. The Functional As-sessment of Cancer Therapy scale: development andvalidation of the general measure. J Clin Oncol. 1993;11:570–9.

24. Temel JS, Greer JA, Muzikansky A, et al. Early palliativecare for patients with metastatic non-small-cell lung can-cer. N Engl J Med. 2010;363:733–42.

25. Trotti A, Bentzen SM. The need for adverse effects re-porting standards in oncology clinical trials. J Clin Oncol.2004;22:19–22.

26. Stiff PJ, Emmanouilides C, Bensinger WI, et al. Paliferminreduces patient-reported mouth and throat soreness andimproves patient functioning in the hematopoietic stem-cell transplantation setting. J Clin Oncol. 2006;24:5186–93.

27. Stiff PJ, Erder H, Bensinger WI, et al. Reliability andvalidity of a patient self-administered daily questionnaireto assess impact of oral mucositis (OM) on pain and dailyfunctioning in patients undergoing autologous hemato-poietic stem cell transplantation (HSCT). Bone MarrowTransplant. 2006;37:393–401.

28. Epstein JB, Beaumont JL, Gwede CK, et al. Longitudinalevaluation of the oral mucositis weekly questionnaire-head and neck cancer, a patient-reported outcomesquestionnaire. Cancer. 2007;109:1914–22.

29. Tomlinson D, Judd P, Hendershot E, Maloney AM, SungL, Establishing literature-based items for an oral mucosi-tis assessment tool in children. J Pediatr Oncol Nurs.2008;25:139–47.

30. Kushner JA, Lawrence HP, Shoval I, et al. Developmentand validation of a Patient-Reported Oral MucositisSymptom (PROMS) scale. J Can Dent Assoc. 2008;74:59.

31. Abernethy AP, Zafar SY, Uronis H, et al. Validation of thePatient Care Monitor (version 2.0): a review of systemassessment instrument for cancer patients. J Pain Symp-tom Manage. 2010;40:545–58.

32. Basch E, Goldfarb S. Electronic patient-reported out-comes for collecting sensitive information from patients.

J Support Oncol. 2009;7:98–9.