The symptom burden of cancer: Evidence for a core set of cancer-related and treatment-related symptoms from the Eastern Cooperative Oncology Group Symptom Outcomes and Practice Patterns study

Authors


  • We thank Jeanie F. Woodruff, BS, ELS, for editorial assistance. The Eastern Cooperative Oncology Group contributed the Symptom Outcomes and Practice Patterns study data.

Abstract

BACKGROUND

A set of common cancer-related and treatment-related symptoms has been proposed for quality of care assessment and clinical research. Using data from a large, multicenter, prospective study, the authors assessed the effects of disease site and stage on the percentages of patients rating these proposed symptoms as moderate to severe.

METHODS

The severity of 13 symptoms proposed to represent “core” oncology symptoms was rated by 3106 ambulatory patients with cancer of the breast, prostate, colon/rectum, or lung, regardless of disease stage or phase of care; 2801 patients (90%) repeated the assessment 4 to 5 weeks later.

RESULTS

At the time of the initial assessment, approximately 33% of the patients reported ≥ 3 symptoms in the moderate-to-severe range; 11 of the 13 symptoms were rated as moderate to severe by at least 10% of all patients and 6 were rated as moderate to severe by at least 20% of those receiving active treatment. Fatigue/tiredness was the most severe symptom, followed by disturbed sleep, pain, dry mouth, and numbness/tingling. More patients with lung cancer and patients receiving active treatment reported moderate to severe symptoms. Percentages of symptomatic patients increased by disease stage, less adequate response to therapy, and declining Eastern Cooperative Oncology Group performance status. The percentages of patients reporting moderate to severe symptoms were stable across both assessments.

CONCLUSIONS

The results of the current study support a core set of moderate to severe symptoms that are common across outpatients with solid tumors, that can guide consideration of progression-free survival as a trial outcome, and that should be considered in clinical care and in assessments of quality of care and treatment benefit. Cancer 2013;119:4333–4340. © 2013 American Cancer Society.

INTRODUCTION

The control of disease-related and treatment-related symptoms is an integral component of cancer care, and symptom reduction is essential for patients with advanced disease. Substantial research has described the prevalence and severity of cancer-related pain and has identified factors associated with its expression and the adequacy of its treatment.[1-3] However, patients with cancer often rate other symptoms, including fatigue, poor appetite, and disrupted sleep, as equally or even more distressing[4, 5]; reducing or delaying the onset of such symptoms would be a significant treatment benefit. Similarly, progression-free survival is becoming a standard endpoint in oncology, yet the critical association between it and symptoms is difficult to estimate because the symptomatic status of patients who develop disease progression has not to our knowledge been well defined.

Studies have identified various symptoms as commonly occurring and attributable both to the disease itself and to the toxicities of cancer therapy,[6, 7] and there is great interest in having a set of “core” symptoms to guide oncology research[6] and clinical care.[8] A panel of symptom assessment experts from the Center for Medical Technology Policy (CMTP) has developed recommendations for a core set of common cancer-related and treatment-related symptoms to be included in adult oncology clinical effectiveness studies, especially studies of patients with advanced disease or who are undergoing aggressive therapy.[9] The concept of a core set of symptoms is supported by systematic reviews of studies with large patient samples (both inpatients and outpatients) and various disease sites.[4, 5] The CMTP panel identified 12 common symptoms that affect patients to a meaningful degree: anorexia, anxiety, constipation, depression, diarrhea, dyspnea, fatigue, insomnia, nausea, pain, neuropathy, and vomiting.

In 2002, a National Institutes of Health State-of-the-Science panel[10] called for more data regarding the occurrence and treatment of pain, depression, and fatigue in adequately powered prospective studies. In response to this call, the Eastern Cooperative Oncology Group (ECOG) designed the large, multisite Symptom Outcomes and Practice Patterns (SOAPP) study to characterize common treatment-related and disease-related symptoms in oncology outpatient practice and to identify the relative influence of factors such as disease site, disease stage, disease status, active treatment, and performance status on patients' symptomatic status. Outpatients with cancers of high prevalence (breast, colorectal, prostate, and lung) were selected for the current study.

We analyzed SOAPP data to characterize the prevalence of the core symptoms proposed by the CMTP panel at moderate to severe levels across the 4 disease sites and to examine clinical characteristics that influence symptom severity. We hypothesized that symptoms would be substantially more severe in patients with progressive disease, later-stage cancer, or worse performance status. We also expected that, due to treatment toxicities, symptoms would be worse for patients currently receiving cancer therapy and that more patients with lung cancer would report severe symptoms compared with patients in the other 3 disease groups.

MATERIALS AND METHODS

Participants

From March 3, 2006 to May 19, 2008, a total of 3123 outpatients with invasive breast, colorectal, prostate, or lung cancer were enrolled in the SOAPP study at any point in the trajectory of their care. Patients were registered at 38 ECOG-affiliated institutions, including 6 academic sites and 32 community clinics. To reduce selection bias, each site devised an ECOG-approved recruitment algorithm that was not biased toward symptom management issues. Eligible patients were aged ≥ 18 years, willing to complete the follow-up survey, and judged by the study screener to have adequate cognitive function to complete study surveys.

The protocol was approved by the Institutional Review Boards at each registering institution. The protocol and case report forms are accessible at ecogsoapp.org.

Procedures

Patients were recruited at the time of check-in for their clinic appointments and provided written informed consent at the time of enrollment. Basic clinical and demographic information, including disease stage, current treatment, disease status, and ECOG performance status (ECOG PS), was then collected by study staff, before the clinician's visit. ECOG PS is a physician-rated measure of functional ability, ranging from 0 (fully functional) to 4 (completely disabled).[11]

Symptom assessment was completed at the time of the initial visit and at a follow-up visit 28 to 35 days later. Patients were asked to complete all questionnaire items according to their experience during the preceding 24 hours. Patients who were too ill to complete the follow-up questionnaire were given the option to mail the forms to the treating clinic by day 42 after the initial visit.

Symptom Assessment

Patients reported symptom intensity using an expanded version of the 19-item M. D. Anderson Symptom Inventory (MDASI).[12] Symptoms are rated “at their worst” in the previous 24 hours on an 11-point numeric scale ranging from 0 (“not present”) to 10 (“as bad as you can imagine”). For the SOAPP study, diarrhea, constipation, sore mouth, rash, hair loss, and coughing were added to the 13 MDASI symptoms. Of these, diarrhea and constipation comprised the 2 remaining proposed core symptoms from the CMTP panel.

In this report, we focused on 13 of the 19 symptoms assessed in the SOAPP study: the 12 symptoms suggested by the CMTP group[9] (fatigue [MDASI fatigue/tiredness], insomnia [MDASI disturbed sleep], pain, anxiety [MDASI distress], neuropathy [MDASI numbness/tingling], dyspnea [MDASI shortness of breath], anorexia [MDASI lack of appetite], depression [MDASI sadness], constipation, diarrhea, nausea, and vomiting) plus xerostomia (MDASI dry mouth), which systematic reviews reported as highly prevalent.[4, 5]

Statistical Analysis

Descriptive statistics present estimates of the prevalence of moderate to severe symptoms across disease sites (breast, colorectal, prostate, and lung) and in the presence of other variables, including current therapy (yes or no), ECOG PS (0, 1, or 2-4), disease status (complete remission, partial remission, stable disease, or progressive disease), and disease stage (no evidence of disease [NED], locoregional, metastatic, or locoregional/metastatic). Chi-square tests were used for subgroup comparisons. The chi-square test for trend was used to test the overall trend of change in the percentage of moderate to severe symptoms by change in disease stage, disease status, and ECOG PS. No adjustment was made for multiple comparisons. All P values were 2-sided and a P value < .05 was considered to be statistically significant. Stata statistical software (version 11.2) was used for all data analysis.[13]

We defined a moderate to severe symptom as one rated ≥ 5 on the MDASI's scale of 0 to 10, on the basis of results from various studies, primarily of cancer-related pain. “Pain at its worst” rated ≥ 5 causes disproportionately greater interference with function, as measured in patients with cancer[14, 15] and community samples.[16] Although the cutpoint used to delineate “moderate to severe” has varied somewhat across studies and even among symptoms (ranging from 4 to 6), we reasoned that using 5 across the board would be a simplifying compromise that offered analytical consistency across all symptoms.

RESULTS

Of the 3123 patients enrolled, 3106 completed the initial MDASI and were included in the current analysis. The sample included 1544 patients with breast cancer (50%), 718 patients with colorectal cancer (23%), 320 patients with prostate cancer (10%), and 524 patients with lung cancer (17%) (Table 1). The median age was 61 years. The sample was predominantly white (86%), non-Hispanic (90%), and female (70%). The median time from the initial cancer diagnosis to study registration was 1.2 years; 2299 patients (74%) were receiving cancer treatment. At the follow-up assessment, 2801 patients completed the MDASI; retention rates were 90% for patients with breast cancer, 93% for patients with colorectal cancer, 90% for patients with prostate cancer, and 87% for patients with lung cancer.

Table 1. Patient Demographic and Disease Characteristics (N = 3106)
VariableNo.%
  1. Abbreviations: CCOP, community clinical oncology program; ECOG PS, Eastern Cooperative Oncology Group performance status; MB-CCOP, minority-based community clinical oncology program; NED, no evidence of disease.

Median age (range), y6118-93
Median y since cancer diagnosis (range) (n = 51 missing data)1.20.0-52.3
Sex  
Women217069.9
Men93630.1
Race  
White264886.3
Black36411.9
Others571.8
Missing data37 
Ethnicity  
Hispanic28510.0
Non-Hispanic257290.0
Missing data249 
ECOG PS  
0175556.8
1110535.8
2-42317.5
Missing data15 
Disease site  
Breast154449.7
Colorectal71823.1
Prostate32010.3
Lung52416.9
Disease status  
Complete remission115737.5
Partial remission1474.8
Stable disease133643.3
Progressive disease44614.5
Missing data20 
Disease stage  
NED133243.0
Locoregional58919.0
Metastatic95931.0
Locoregional plus metastatic2157.0
Missing data11 
Patient currently receive therapy  
No80726.0
Yes229974.0
Prior systemic therapy  
No119538.5
Yes191061.5
Missing data1 
Prior radiotherapy  
No178257.9
Yes129742.1
Missing data27 
Type of institution  
Academic3039.8
CCOP221471.3
MB-CCOP58919.0

Symptom Severity at Initial Assessment

By Disease Site

By disease site, the percentages of patients rating ≥ 3 core symptoms as moderate to severe ranged from 28% to 47% (Table 2). Overall, 11 of the 13 symptoms were rated as moderate to severe by ≥ 10% of all patients (Table 3). Vomiting was rated as moderate or severe by < 4% of patients. Fatigue/tiredness was the most prevalent moderate to severe symptom in each of the cancer groups, ranging from 31% in patients with breast cancer to 46% in those with lung cancer.

Table 2. Percentages of Patients Reporting Moderate to Severe Symptoms at Initial Assessment by Number of Symptoms Reported
No. of Moderate to Severe Symptoms ReportedaDisease SitebCurrent Therapyc
Breast n = 1544Colorectal n = 718Prostate n = 320Lung n = 524No n = 807Yes n = 2299
  1. a

    Categories are not mutually exclusive; columns may total >100%.

  2. b

    Compared with the other disease types (chi-square test), a greater percentage of patients with lung cancer reported 1) ≥3 moderate to severe symptoms (P <.001 for all 3 pairwise comparisons); 2) ≥5 moderate to severe symptoms (P <.001 for all 3 pairwise comparisons); and 3) ≥7 moderate to severe symptoms (P ≤.001for all 3 pairwise comparisons).

  3. c

    Compared with patients not undergoing treatment (chi-square test), a greater percentage of patients currently receiving treatment reported 1) ≥3 moderate to severe symptoms (P <.001); 2) ≥5 moderate to severe symptoms (P <.001); and 3) ≥7 moderate to severe symptoms (P <.005).

047.043.743.130.551.240.2
≥330.131.627.847.127.235.2
≥517.018.515.328.114.520.7
≥78.59.17.214.97.110.5
Table 3. Percentages of Patients Reporting Moderate to Severe Symptoms at Initial Assessment by Disease Site and Current Therapy
SymptomaDisease SitebCurrent TherapycTotal n = 3106
Breast n = 1544Colorectal n = 718Prostate n = 320Lung n = 524No n = 807Yes n = 2299
  1. a

    The 5 most prevalent symptoms for each column are presented in bold type. Of the 3106 patients who completed the initial assessment, the percentages of patients with missing values for a given symptom ranged from 0.7% to 2.4%.

  2. b

    A higher percentage of patients with lung cancer reported moderate to severe fatigue/tiredness, disturbed sleep, pain, dry mouth, distress, shortness of breath, lack of appetite, sadness, constipation, and nausea compared with other patients combined (all P <.05, chi-square test).

  3. c

    A higher percentage of patients receiving active therapy reported moderate to severe fatigue/tiredness, disturbed sleep, nausea, lack of appetite, dry mouth, sadness, vomiting, diarrhea, and constipation compared with patients not currently receiving therapy (all P <.05, chi-square test).

Fatigue/tiredness31.332.935.546.126.737.434.6
Disturbed sleep27.523.025.631.022.128.526.9
Pain18.518.517.524.317.719.919.4
Dry mouth19.115.715.926.414.420.919.2
Distress18.917.913.523.416.619.718.9
Numbness/tingling16.925.815.818.417.119.819.1
Shortness of breath10.411.416.133.914.315.515.2
Lack of appetite11.917.311.127.111.717.015.6
Sadness16.114.114.520.913.417.316.3
Constipation11.411.012.419.99.514.112.9
Diarrhea8.915.88.89.47.911.510.5
Nausea7.010.04.710.75.09.28.1
Vomiting3.53.51.63.71.83.93.4

Patients with lung cancer were the most symptomatic group, with nearly 50% rating ≥ 3 symptoms as moderate to severe (Table 2). Compared with patients in the other disease groups, a higher percentage of patients with lung cancer reported moderate to severe fatigue/tiredness, disturbed sleep, pain, dry mouth, distress, shortness of breath, lack of appetite, sadness, constipation, and nausea (all P < .05) (Table 3). In addition, 28% of patients with lung cancer rated at least 5 of the 13 symptoms as moderate or severe, compared with 17% of patients with breast, 19% of patients with colorectal, and 15% of patients with prostate cancer (P < .001 for all 3 pairwise comparisons).

By Treatment Status

More than 35% of patients currently receiving treatment reported ≥ 3 moderate to severe symptoms (Table 2). Grouped by treatment status, > 20% of patients reported moderate to severe fatigue/tiredness and disturbed sleep; percentages were significantly higher for patients receiving treatment (both P < .001) (Table 3). Significantly more patients receiving treatment also reported moderate to severe nausea, lack of appetite, dry mouth, sadness, vomiting, diarrhea, and constipation (all P < .05). Prior systemic therapy was associated with increased pain (16.1% vs 21.4%; P < .001) and numbness/tingling (15.1% vs 21.6%; P < .001).

By Disease Stage, Disease Status, and ECOG PS

Overall, the percentage of patients reporting each symptom as moderate to severe increased as the disease stage worsened (Table 4). However, > 20% of patients with NED reported moderate to severe fatigue/tiredness and disturbed sleep.

Table 4. Percentages of Patients Reporting Moderate to Severe Symptoms at Initial Assessment by Disease Stage, Disease Status, and ECOG PS{TC}
SymptomaDisease Stageb, cDisease Statusb, dECOG PSb, e
NED n = 1332LR n = 589Mets n = 959LR + Mets n = 215CR n = 1157PR n = 147SD n = 1336PD n = 4460 n = 17551 n = 11052-4 n = 231
  1. Abbreviations: CR, complete response; ECOG PS, Eastern Cooperative Oncology Group performance status; LR, locoregional; Mets, metastatic disease; NED, no evidence of disease; PD, progressive disease; PR, partial response; SD, stable disease.

  2. a

    The 5 most prevalent symptoms for each column are presented in bold type. Of the 3106 patients who completed the initial assessment, the percentages of patients with missing values for a given symptom ranged from 0.7% to 2.4%.

  3. b

    The chi-square test for trend demonstrated an increasing percentage of moderate to severe symptoms for all 13 symptoms by increasing disease stage (all P <.01), disease status (all P <.001), and ECOG PS (all P <.0001).

  4. c

    Compared with patients with NED (chi-square test), 1) a higher percentage of patients with metastatic disease (metastatic and LR/metastatic categories combined) reported moderate to severe levels for all 13 symptoms (all P <.05) and 2) a higher percentage of patients with LR disease reported having moderate to severe fatigue/tiredness, nausea, disturbed sleep, distress, shortness of breath, lack of appetite, dry mouth, sadness, vomiting, diarrhea, and constipation (all P <.05).

  5. d

    Compared with patients who achieved CR (chi-square test), 1) a higher percentage of patients with SD or PD reported having moderate to severe levels of all 13 symptoms (all P <.001) and 2) a higher percentage of patients who achieved a PR reported moderate to severe fatigue/tiredness, nausea, shortness of breath, lack of appetite, dry mouth, vomiting, and numbness/tingling (all P <.05).

  6. e

    For all 13 symptoms, a higher percentage of patients with an ECOG PS >0 had moderate to severe symptoms compared with patients with an ECOG PS of 0 (P <.001 for an ECOG PS of 2 to 4 vs 0 and an ECOG PS of 1 vs 0, chi-square test). Compared with patients with an ECOG PS of 1, a higher percentage of patients with an ECOG PS of 2 to 4 had moderate to severe pain, fatigue/tiredness, nausea, disturbed sleep, shortness of breath, lack of appetite, dry mouth, numbness/tingling, and constipation (all P <.05, chi-square test).

Fatigue/tiredness24.738.442.352.923.944.039.645.423.944.867.7
Disturbed sleep23.227.630.530.822.828.828.830.620.633.641.9
Pain14.016.625.731.513.917.920.929.611.526.246.0
Dry mouth14.720.722.628.814.621.519.928.612.126.139.4
Distress16.220.719.229.915.318.920.522.812.825.531.7
Numbness/tingling15.614.625.524.315.422.819.626.613.225.035.2
Shortness of breath9.718.418.426.39.218.517.722.17.822.734.2
Lack of appetite8.518.921.425.88.016.718.626.08.122.340.3
Sadness14.017.917.222.013.615.117.121.411.221.927.6
Constipation8.813.916.220.88.411.115.318.18.317.026.9
Diarrhea8.611.511.316.78.18.212.511.87.214.914.5
Nausea5.48.411.111.25.49.69.510.54.611.716.8
Vomiting2.03.84.93.31.94.83.85.01.85.15.7

A similar pattern of increasing symptom severity was noted when patients were characterized by disease status and ECOG PS (Table 4). Higher percentages of patients with stable or progressive disease reported moderate to severe levels for all symptoms compared with patients who achieved a complete response (all P < .05). Patients with an ECOG PS ≥ 2 were highly symptomatic. The percentages of patients with an ECOG PS of 1 reporting moderate to severe symptoms were significantly greater than the percentages of patients with an ECOG PS of 0 reporting moderate to severe symptoms (P < .001 for all 13 symptoms).

Symptom Severity at Follow-Up

In the aggregate, severity ratings made by patients at the follow-up visit were similar to those made at the time of the initial assessment, even though many individual patients did demonstrate changes in symptom ratings between the 2 visits. For all disease sites, the change in the percentages of patients reporting moderate to severe symptoms was within 5.4%. The largest discrepancy between baseline and follow-up was an increase in patients with lung cancer reporting moderate to severe pain. The majority of the top 5 symptoms remained consistent over time, in all patients as a group and by disease site, current therapy, disease stage, disease status, and ECOG PS (Table 5).

Table 5. Percentages of Patients Reporting Moderate to Severe Symptoms at Follow-Up Assessment
SymptomaDisease SiteCurrent TherapyDisease StageDisease StatusECOG PS 
Breast n = 1389Colorectal n = 666Prostate n = 289Lung n = 457No n = 704Yes n = 2097NED n = 1211LR n = 520Mets n = 867LR + Mets n = 193CR n = 1053PR n = 134SD n = 1216PD n = 3800 n = 16101 n = 9842-4 n = 194Total n = 2801
  1. Abbreviations: CR, complete response; ECOG PS, Eastern Cooperative Oncology Group performance status; LR, locoregional; Mets, metastatic disease; NED, no evidence of disease; PD, progressive disease; PR, partial response; SD, stable disease.

  2. a

    The 5 most prevalent symptoms for each column are presented in bold type. Of the 2801 patients who completed the follow-up assessment, the percentages of patients with missing values for a given symptom ranged from 1.1% to 3.6%.

Fatigue/tiredness34.531.831.150.627.439.028.139.143.147.327.545.538.947.628.044.063.036.1
Disturbed sleep23.222.921.728.421.324.721.427.923.229.621.128.625.224.919.128.239.623.8
Pain19.216.017.029.716.221.214.321.724.828.113.523.123.226.113.825.342.219.9
Dry mouth16.315.515.322.413.818.113.919.018.325.113.222.617.324.412.122.927.117.0
Distress18.918.618.320.117.219.516.819.520.126.516.220.319.324.314.423.035.419.0
Numbness/tingling16.825.616.721.317.920.216.917.623.326.216.422.620.026.514.524.934.919.6
Shortness of breath11.612.114.932.414.715.711.018.818.022.510.417.417.721.610.120.832.115.5
Lack of appetite12.815.39.425.711.416.49.418.419.423.78.721.117.523.19.620.633.315.1
Sadness16.315.613.919.713.917.314.418.216.126.614.017.916.421.812.420.728.916.5
Constipation12.810.910.520.49.514.610.415.215.118.79.912.815.316.910.416.223.413.3
Diarrhea7.915.76.98.47.610.57.512.111.58.96.611.211.511.97.913.08.99.7
Nausea7.910.32.411.34.69.85.711.39.413.76.09.79.710.76.710.512.68.5
Vomiting3.04.71.44.92.83.82.25.24.44.32.14.53.86.41.95.46.83.6

DISCUSSION

The results of the current study illustrate that patients with cancer are at risk for experiencing multiple moderate to severe symptoms simultaneously and support the use of routine clinical screening for a core set of symptoms in addition to pain, especially for patients receiving active treatment, those with diminished ECOG PS, or those with metastatic disease.

Greater than 25% (prostate cancer) to nearly 50% (lung cancer) of the sample rated ≥ 3 core symptoms as moderate to severe. Fatigue/tiredness was rated as moderate to severe by the greatest percentages of patients across the 4 disease groups. Disturbed sleep and pain also were prevalent. Dry mouth, which was not among the core symptoms proposed by the CMTP,[9] was among the most frequent moderate to severe symptoms reported in the current study. Numbness/tingling was second in prevalence for patients with colorectal cancer, and most likely is a product of the neurotoxic therapies used to treat this cancer type. Although moderate to severe nausea was reported by >10% of patients with an ECOG PS of  > 0 or metastatic disease, vomiting was rarely reported. Overall, 11 of the 12 CMTP symptoms (excluding vomiting) were rated as moderate or severe by ≥ 10% of at least 1 of the disease groups, suggesting that they are candidates for routine clinic screening.

Although individual patients demonstrated an improvement or worsening of their symptoms between the initial and follow-up assessments, the overall percentages of patients reporting moderate to severe symptoms for each disease group and by disease stage, disease status, and ECOG PS were quite stable between the 2 assessments. The stability of the ratings supports the representativeness of the selected symptoms and their ordering by severity.

Expectations for associations between greater symptom burden and more advanced disease, poor response to therapy, poor ECOG PS (≥ 2), and lung cancer were supported by the data. Less expected was the substantial symptomatic difference noted between patients with an ECOG PS of 1 (ambulatory, able to perform light work) and those with an ECOG PS of 0 (fully active). At least 20% of patients with an ECOG PS of 1 reported moderate to severe levels for 9 of the 13 symptoms, as opposed to only 2 symptoms (fatigue/tiredness and disturbed sleep) for those with an ECOG PS of 0. This suggests that performance status may need to be included as a stratification or eligibility factor in addition to disease status and stage in trials in which change in symptom status is a variable of interest. In contrast to results from previous studies,[1, 17] we found no differences in symptom severity by patient sex (data not shown).

Our use of a fixed cutpoint of ≥ 5 to define a moderate to severe symptom is a potential limitation. Others have suggested that different symptoms may become moderate to severe at different ratings on a scale of 0 to 10.[18] Practice guidelines for the management of cancer-related pain, fatigue, and distress[19] suggest that these symptoms are moderate when rated as 4. It is not our intention to indicate that symptoms rated < 5 are not clinically significant and do not warrant clinical care; we chose a cutpoint of 5 because it was not likely to exclude those patients with significant symptoms. More research on cutpoints for severity ratings across symptoms is needed.

Second, we did not factor in comorbid conditions (such as arthritis or cardiovascular disease) that may have contributed to symptom burden. However, it is likely that any such comorbid conditions would have equally affected the cancer-free patients (those with NED) in this sample, > 10% of whom rated 7 of the 13 symptoms as moderate or severe (Table 4). Similarly, we purposely made no attempt to attribute symptoms to treatment or disease, because making such attribution is difficult at best and unreliable at worst.[6, 20]

Finally, we restricted our investigation to the 12 core symptoms recommended by the CMTP expert panel[9] and 1 additional symptom (dry mouth) that was shown to be prevalent in 2 extensive systematic reviews.[4, 5] The SOAPP data were generally consistent with the systematic reviews and supported the CMTP core symptom list with the exception of vomiting, which was rarely reported, even by the patients with the most advanced disease. In addition, the high percentage of patients reporting moderate to severe dry mouth suggests that it is a candidate for inclusion in a core list. MDASI symptoms not identified as prevalent by either the CMTP panel or the systematic reviews, and therefore not included in the current analysis, were drowsiness and difficulty remembering. These symptoms were rated as moderate to severe by 22.9% and 17.2% of patients, respectively, in the SOAPP sample, indicating that they also may warrant inclusion in a core list. It is possible that other symptoms not contemplated here could be prevalent at moderate to severe levels in patients with cancer. Future studies may resolve such discrepancies; notwithstanding, we believe that the 12 symptoms corroborated in the current study form a reliable representation of symptom burden that would aid quality-of-care evaluations and clarify the patient's perspective on the symptomatic consequences of disease progression.

It is increasingly routine to have patients rate pain on a scale of 0 to 10 at clinic visits. Adding a limited number of core symptoms would not appreciably increase patient burden, and doing so might capture symptoms that are even more distressing than pain and could provide clinical teams with more robust information with which to understand the patient within the context of cancer care, especially if high levels of symptoms are documented and trigger an evaluation and management plan.[8]

FUNDING SUPPORT

Supported by grants from the National Cancer Institute/National Institutes of Health, including U10 grants CA37403, CA15488, CA13650, CA35412, and CA063847 to the Eastern Cooperative Oncology Group; R01 CA026582 to Dr. Cleeland; and MD Anderson Cancer Center Support Grant P30 CA016672. Neither the National Cancer Institute nor the National Institutes of Health had any role in the study design, data collection, analysis, interpretation, or preparation of this article.

CONFLICT OF INTEREST DISCLOSURES

The authors made no disclosures.

Ancillary