Symptom clusters of gastrointestinal cancer patients ... - mascc

3 downloads 0 Views 165KB Size Report
sis for the entirety of gastrointestinal cancer patients' radio- therapy treatments. ...... evaluate CINV could be applied to agents of any emetogenic potential [18].
Support Care Cancer (2015) 23:2589–2598 DOI 10.1007/s00520-015-2617-9

ORIGINAL ARTICLE

Symptom clusters of gastrointestinal cancer patients undergoing radiotherapy using the Functional Living Index—Emesis (FLIE) quality-of-life tool Michael Poon & Kristopher Dennis & Carlo DeAngelis & Hans Chung & Jordan Stinson & Liying Zhang & Gillian Bedard & Erin Wong & Marko Popovic & Nicholas Lao & Natalie Pulenzas & Shun Wong & Paul Cheon & Edward Chow

Received: 13 August 2014 / Accepted: 13 January 2015 / Published online: 27 January 2015 # Springer-Verlag Berlin Heidelberg 2015

Abstract Objective The Functional Living Index—Emesis (FLIE) instrument is a validated nausea and vomiting specific quality of life (QOL) tool originally created as a 3-day test of the impact of chemotherapy-induced nausea and vomiting on cancer patients’ daily life. The primary objective of the present study was to retrospectively explore the use of the FLIE from data obtained in a previously published study of patients with gastrointestinal radiation-induced nausea and vomiting (RINV) and compare the extracted symptom clusters on a weekly basis for the entirety of gastrointestinal cancer patients’ radiotherapy treatments. Methods QOL was assessed on a weekly basis using the 18item FLIE questionnaire for patients’ radiotherapy treatments. A principal component analysis with varimax rotation was performed at each visit. The internal consistency and reliability of the derived clusters was assessed with Cronbach’s alpha. Robust relationship and correlation among symptoms was displayed with biplot graphics. Results A total of 460 FLIE assessments were completed for the 86 gastrointestinal patients who underwent radiotherapy.

Two components were consistently identified except for week 5 where only one component was identified. Component 1 contained the items BQ10–Q18^ which included all vomiting items. Component 2 included all nausea items from BQ1 to Q9^. All the variables were well accounted for by two components for most weeks of treatment with excellent internal consistency. Biplots indicate that the two symptom clusters were evident at each week, with the exception of the first week of treatment. Strong correlations were seen between the effect of nausea on patients’ ability to make meals, patients’ ability to do tasks within the home, and patients’ willingness to spend time with family and friends. Conclusion The high internal consistency at all timepoints indicates that the FLIE QOL instrument is useful for the RINV population. Keywords Radiotherapy-induced nausea and vomiting . Symptom cluster . Functional Living Index—Emesis

Introduction M. Poon : C. DeAngelis : H. Chung : J. Stinson : L. Zhang : G. Bedard : E. Wong : M. Popovic : N. Lao : N. Pulenzas : S. Wong : P. Cheon : E. Chow (*) Department of Radiation Oncology, Odette Cancer Centre, Sunnybrook Health Sciences Centre, University of Toronto, 2075 Bayview Avenue, Toronto, Ontario, Canada M4N 3M5 e-mail: [email protected] K. Dennis Division of Radiation Oncology, Ottawa Hospital Research Institute, University of Ottawa, Ottawa, Ontario, Canada C. DeAngelis Leslie Dan Faculty of Pharmacy, University of Toronto, Toronto, Ontario, Canada

Nausea and vomiting are common side effects experienced by cancer patients that result from both their disease and antineoplastic therapies [1]. A review of nausea and vomiting in cancer patients found that rates of these symptoms range from 4 to 44 % prior to treatment [2]. In the radiation setting, depending on the primary cancer site and the stage of disease, this value increases to an estimated 40–80 % of patients developing radiotherapy-induced nausea and vomiting (RINV) [3, 4]. This RINV can interfere with patients’ quality of life (QOL) and lead to treatment delay or refusal [1]. Therefore, it is necessary to be proactive about managing these symptoms and reducing their occurrence.

2590

The current body of literature into RINV is dwarfed by the amount of research that has been invested into chemotherapy (CT)-induced nausea and vomiting (CINV). The causes of RINV are not fully understood; however, certain factors have been identified that are believed to increase the risk of this effect. The most significant factor is thought to be the anatomic area being irradiated. However, other patient and treatment characteristics, including fractionation, single and total treatment dose, volume of small bowel radiated, age, gender, psychological state, tumor stage, and whether the patient has concurrent or recent chemotherapy, should not be overlooked when assessing a patient’s emetogenic risk [5]. Treatment to the upper abdomen, such as that for many gastrointestinal cancers, carries an estimated 60–90 % risk of emesis, while treatment to the pelvis, such as that for many lower gastrointestinal cancers, carries an estimated 30–60 % risk [6]. Consequently, patients with planned gastrointestinal radiotherapy–treatment are ideal to assess an instrument’s ability to score nausea and vomiting effect on QOL. The Functional Living Index—Emesis (FLIE) instrument is a nausea- and vomiting-specific quality of life (QOL) tool originally created as a 3-day test of the impact of CINV on cancer patients’ daily life. It was validated using Pearson and Cronbach alpha correlations and the nausea factor of the Functional Living Index—Cancer (FLIC) during a clinical trial by Lindley et al. [7]. The FLIE is modeled after the FLIC’s questions, length, and format. However, it is designed to address the impact of nausea and vomiting on physical activities, social and emotional function, and ability to enjoy meals [7]. This 18-item tool contains nine items for nausea and nine items for vomiting, with a possible summary score ranging from 18 to 126. Lower scores indicate an increased negative impact on a patient's functional living due to nausea and vomiting [7]. Since its validation, the FLIE has been used to evaluate QOL following administration of antiemetics such as ondansetron, granisetron, and aprepitant [8–11]. Since there is no nausea- and vomitingspecific QOL instrument for radiotherapy and given the similar patient population, the FLIE was selected to measure the functional impact of radiotherapy in the present study. The term Bsymptom cluster^ was first coined by Dodd et al. in 2001 in their research into pain, fatigue, and sleep disturbances [12]. It is defined by at least two or more interrelated symptoms which occur concurrently in a stable group that is relatively independent of other clusters. This similarity in patient reporting is thought to potentially reveal specific underlying dimensions or mechanisms. While these symptoms may or may not have a singular underlying cause, to be considered clustered, the symptoms or responses must have a stronger relationship with in-group symptoms than symptoms in other clusters. The primary objectives of the present study were to retrospectively explore the use of the FLIE from data obtained in a previously published single-institution prospective observational symptom study of patients with GI cancers undergoing

Support Care Cancer (2015) 23:2589–2598

radiotherapy and at risk of RINV with or without chemotherapy and compare the extracted clusters of functions that could be impaired on a weekly basis during patients’ radiotherapy treatments [13].

Methods All patients from the previous study were considered eligible as long as they were aged 18 years or older at the time of enrolment, had at least a Karnofsky Performance Status of 40, and had a histologically, cytologically, or radiologically proven gastrointestinal tumor. Patients who had received prior radiotherapy that targeted intra-abdominal and/or intra-pelvic structures, radiotherapy to the brain during the 7 days prior to the day of study radiotherapy commencement, or were planned to receive radiotherapy to the brain during the study period or 7 days following the study radiotherapy period were ineligible for this study. All radiotherapy, chemotherapy treatment plans, and antiemetic therapies were left to the discretion of the patient’s treating oncologists. QOL was assessed on a weekly basis using the FLIE questionnaire, beginning at baseline before the first treatment of each patient, for the entirety of each patient’s treatment. Every 7th day, the patient met with a research assistant, in person, to complete the QOL assessment. If the planned in-person meeting did not occur for any reason, the patient was contacted via telephone to collect the information on the same day. If this was not possible, the information was collected in person when the patient attended the next radiation treatment. This occurred until the patient’s last day of treatment, when the patient would complete their final during-treatment QOL assessment. The patient was then called on the 7th day after the final treatment for a final follow-up QOL assessment. The FLIE is an 18-item questionnaire that contains nine items for nausea and nine items for vomiting (Appendix). The majority of items are scored on a scale from 1 to 7, with 1 indicating Bnone^ and 7 indicating Ba great deal^. Items 3, 6, 11, 15, and 18 are scored in the opposite manner, with 1 indicating Ba great deal^ and 7 indicating Bnone^. Statistical analysis To examine whether any interrelationships existed between symptoms and functional scales (FLIE 18 items), a principal component analysis (PCA) with varimax rotation was performed on the 18 FLIE items at each visit. The PCA transforms a number of observed variables into a smaller number of variables (called principal components). The first principal component or Bcluster^ accounts for as much of the variability in the data as possible. The total variance should be 18 as they have been standardized to have a standard deviation (SD) of 1; each symptom contributes one unit of variance to total

Support Care Cancer (2015) 23:2589–2598

variance. The amount of variance that is captured by a given component is named an Beigenvalue^. Significant principal components were selected if they possessed an eigenvalue greater than 1.0, and each component accounted for at least 10 % of total variation. The highest factor loading score predicted the assignment of individual symptoms to an independent factor. The internal consistency and reliability of the derived clusters was assessed with Cronbach’s alpha. Robust relationships and correlations among symptoms were displayed with the biplot graphic. The longer the length and closer together the arrows were, the higher the correlation between symptoms. All analyses were conducted by Statistical Analysis Software (SAS version 9.3).

Results Eighty-six patients who were planned to receive curative or palliative abdominal and/or pelvic radiotherapy alone or with concurrent chemotherapy had sufficient FLIE data to be included within our analyses. A total of 460 FLIE QOL assessments were completed. Of those enrolled, five patients did not receive their planned treatment and only completed a baseline evaluation. Patients’ treatment lengths varied from less than a week to 7 weeks, and the number of QOL assessments completed on a given week varied from 86 at baseline to 7 on week 6. Table 1 shows the demographics for those patients who received radiation. The mean, SD, and minimum to maximum were calculated for FLIE 1–18 items at baseline, weeks 1–6, end week, and follow-up week (Table 2). The PCA statistical approach was performed with varimax rotation on the 18 items of the FLIE. Because of the limited number of patients requiring a week 6 QOL assessment, we will not include week 6 in the symptom cluster analysis. Baseline symptom clusters The PCA extracted two components with eigenvalues greater than 1.0 and proportion of variance ≥10 %. Overall, these two components explain 90 % of the total variance, and each component explains more than 10 % of the variance. Component 1 consists of 69 % (12.4 units) of the total variance and was predicted by the highest factor loading score. The component contained the items BQ10–Q18^ which included all vomiting items. Component 2 consists of 21 % (3.9 units) of the total variance, including all nausea items from BQ1 to Q9^. All of the variables are well accounted for by two components, with final communality estimates ranging from 0.72 for Q6 to 0.98 for Q11, Q12, Q15, Q17, and Q18. The internal reliabilities of the two clusters were 0.96 for both, which indicates that the two clusters had excellent internal consistency (Table 3).

2591 Table 1

Demographics of patients who received radiation

Characteristic Number of patients

81

Age (years) Mean Median (range)

62.8 65 (25–92)

Sex Female

36

(45.5 %)

Male

45

(55.5 %)

Primary cancer site, n (percentage) Rectal

33

(40.74 %)

Pancreas Esophagus

14 7

(17.28 %) (8.64 %)

Liver

7

(8.64 %)

Colon Stomach

5 5

(6.17 %) (6.17 %)

Kidney Others

3 7

(3.70 %) (8.64 %)

No Yes Previous RT No

76 5

(93.7 %) (6.3 %)

76

(93.7 %)

Yes Previous chemotherapy No Yes Previous CINV

5

(6.3 %)

61 20

(75.3 %) (24.7 %)

8 12

(40 %) (60 %)

Anxiety disorder

No Yes RT RT duration (days) Median (range) RT emetogenicity risk level Low Moderate Chemotherapy treatment CT planned concurrently No Yes CT received Capecitabine Infusional 5-fluorouracil FU/mitomycin+RT FUCISP CT emetogenicity risk level High Low CT chemotherapy, RT radiotherapy

41(0–57) 13

(16 %)

68

(84 %)

28 53

(34.6 %) (65.4 %)

39 9 2 3

(73.5 %) (17.1 %) (3.8 %) (5.7 %)

3 50

(5.8 %) (94.2 %)

2592

Support Care Cancer (2015) 23:2589–2598

Table 2 Mean, standard deviation, median, and minimum to maximum for all 18-items of the FLIE QOL instrument at baseline, week 1–6, end week, and follow-up week

RINV Q1

RINV Q2

RINV Q3

RINV Q4

RINV Q5

RINV Q6

Visit week

Visit

Number, N

Mean

SD

Median

Min

Max

0

Baseline

86

1.3

1.02

1.0

1.0

6.0

1

Week 1

64

1.9

1.92

1.0

1.0

7.0

2 3

Week 2 Week 3

52 48

1.6 1.8

1.42 1.59

1.0 1.0

1.0 1.0

7.0 7.0

4

Week 4

34

2.0

1.77

1.0

1.0

7.0

5 6

Week 5 Week 6

21 7

2.1 1.9

1.97 1.57

1.0 1.0

1.0 1.0

7.0 5.0

7 8

End week Follow-up week

76 76

2.2 1.7

1.80 1.67

1.0 1.0

1.0 1.0

7.0 7.0

0

Baseline

85

1.2

0.89

1.0

1.0

7.0

1

Week 1

64

1.7

1.69

1.0

1.0

7.0

2 3 4

Week 2 Week 3 Week 4

53 48 34

1.4 1.4 1.7

1.17 1.24 1.58

1.0 1.0 1.0

1.0 1.0 1.0

7.0 7.0 6.0

5

Week 5

20

1.6

1.43

1.0

1.0

6.0

6 7 8

Week 6 End week Follow-up week

7 74 75

1.6 1.8 1.5

1.51 1.79 1.47

1.0 1.0 1.0

1.0 1.0 1.0

5.0 7.0 7.0

0 1 2 3

Baseline Week 1 Week 2 Week 3

86 63 53 48

6.7 6.3 6.5 6.4

1.29 1.87 1.42 1.58

7.0 7.0 7.0 7.0

1.0 1.0 1.0 1.0

7.0 7.0 7.0 7.0

4 5 6

Week 4 Week 5 Week 6

34 21 7

6.6 6.4 6.9

1.10 1.56 0.38

7.0 7.0 7.0

2.0 1.0 6.0

7.0 7.0 7.0

7 8 0 1 2

End week Follow-up week Baseline Week 1 Week 2

73 75 86 64 53

6.2 6.5 1.3 1.7 1.6

1.80 1.47 1.11 1.85 1.56

7.0 7.0 1.0 1.0 1.0

1.0 1.0 1.0 1.0 1.0

7.0 7.0 7.0 7.0 7.0

3 4 5 6 7 8 0 1 2 3 4

Week 3 Week 4 Week 5 Week 6 End week Follow-up week Baseline Week 1 Week 2 Week 3 Week 4

48 34 21 7 74 76 86 64 52 47 34

1.7 2.3 2.1 1.9 2.2 1.6 1.2 1.6 1.4 1.6 1.9

1.43 2.18 2.22 1.86 2.02 1.58 0.79 1.77 1.42 1.39 1.77

1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0

1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0

7.0 7.0 7.0 6.0 7.0 7.0 6.0 7.0 7.0 7.0 7.0

5 6 7 8 0 1 2

Week 5 Week 6 End week Follow-up week Baseline Week 1 Week 2

21 7 74 75 86 64 53

2.0 1.7 1.8 1.6 6.8 6.2 6.6

1.94 1.50 1.73 1.55 0.81 1.90 1.16

1.0 1.0 1.0 1.0 7.0 7.0 7.0

1.0 1.0 1.0 1.0 2.0 1.0 1.0

7.0 5.0 7.0 7.0 7.0 7.0 7.0

Support Care Cancer (2015) 23:2589–2598

2593

Table 2 (continued)

RINV Q7

RINV Q8

RINV Q9

RINV Q10

RINV Q11

Visit week

Visit

Number, N

Mean

SD

Median

Min

Max

3 4

Week 3 Week 4

48 34

6.5 6.5

1.38 1.24

7.0 7.0

1.0 2.0

7.0 7.0

5

Week 5

21

6.4

1.50

7.0

1.0

7.0

6 7

Week 6 End week

7 75

5.6 6.1

2.44 1.86

7.0 7.0

2.0 1.0

7.0 7.0

8

Follow-up week

75

6.5

1.44

7.0

1.0

7.0

0 1

Baseline Week 1

86 64

1.2 1.6

0.87 1.72

1.0 1.0

1.0 1.0

7.0 7.0

2 3

Week 2 Week 3

53 48

1.5 1.5

1.34 1.20

1.0 1.0

1.0 1.0

7.0 7.0

4

Week 4

34

1.7

1.34

1.0

1.0

6.0

5 6

Week 5 Week 6

21 7

1.9 1.3

1.73 0.49

1.0 1.0

1.0 1.0

6.0 2.0

7 8 0

End week Follow-up week Baseline

74 75 86

1.9 1.6 1.2

1.79 1.57 0.82

1.0 1.0 1.0

1.0 1.0 1.0

7.0 7.0 6.0

1 2

Week 1 Week 2

64 53

1.7 1.5

1.78 1.42

1.0 1.0

1.0 1.0

7.0 7.0

3

Week 3

48

1.6

1.29

1.0

1.0

7.0

4 5

Week 4 Week 5

34 21

1.7 2.0

1.34 2.01

1.0 1.0

1.0 1.0

6.0 7.0

6

Week 6

7

1.4

1.13

1.0

1.0

4.0

7 8 0 1 2

End week Follow-up week Baseline Week 1 Week 2

74 75 86 64 53

1.9 1.6 1.2 1.5 1.4

1.81 1.59 1.09 1.52 1.35

1.0 1.0 1.0 1.0 1.0

1.0 1.0 1.0 1.0 1.0

7.0 7.0 7.0 7.0 7.0

3 4 5 6 7

Week 3 Week 4 Week 5 Week 6 End week

48 34 21 7 74

1.3 1.6 1.8 1.0 1.7

1.01 1.42 1.73 0.00 1.76

1.0 1.0 1.0 1.0 1.0

1.0 1.0 1.0 1.0 1.0

7.0 6.0 7.0 1.0 7.0

8 0 1 2 3 4 5 6

Follow-up week Baseline Week 1 Week 2 Week 3 Week 4 Week 5 Week 6

75 86 64 53 48 33 21 7

1.4 1.1 1.6 1.3 1.1 1.2 1.5 1.0

1.41 0.48 1.66 1.18 0.72 0.89 1.25 0.00

1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0

1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0

7.0 4.0 7.0 7.0 6.0 6.0 5.0 1.0

7 8 0 1 2 3 4 5 6

End week Follow-up week Baseline Week 1 Week 2 Week 3 Week 4 Week 5 Week 6

73 76 86 64 53 48 33 20 7

1.5 1.4 6.9 6.6 6.9 6.9 6.9 6.7 7.0

1.33 1.38 0.65 1.22 0.62 0.87 0.38 0.92 0.00

1.0 1.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0

1.0 1.0 1.0 1.0 3.0 1.0 5.0 4.0 7.0

7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0

2594

Support Care Cancer (2015) 23:2589–2598

Table 2 (continued)

RINV Q12

RINV Q13

RINV Q14

RINV Q15

RINV Q16

RINV Q17

Visit week

Visit

Number, N

Mean

SD

Median

Min

Max

7 8

End week Follow-up week

72 76

6.6 6.7

1.39 1.30

7.0 7.0

1.0 1.0

7.0 7.0

0

Baseline

86

1.1

0.54

1.0

1.0

6.0

1 2

Week 1 Week 2

63 53

1.4 1.2

1.44 0.78

1.0 1.0

1.0 1.0

7.0 5.0

3

Week 3

48

1.1

0.87

1.0

1.0

7.0

4 5

Week 4 Week 5

33 20

1.2 1.3

0.71 0.79

1.0 1.0

1.0 1.0

5.0 4.0

6 7

Week 6 End week

7 72

1.0 1.4

0.00 1.35

1.0 1.0

1.0 1.0

1.0 7.0

8

Follow-up week

76

1.3

1.25

1.0

1.0

7.0

0 1

Baseline Week 1

86 64

1.1 1.6

0.58 1.68

1.0 1.0

1.0 1.0

6.0 7.0

2 3 4

Week 2 Week 3 Week 4

53 48 33

1.2 1.1 1.2

0.99 0.87 0.87

1.0 1.0 1.0

1.0 1.0 1.0

7.0 7.0 6.0

5 6

Week 5 Week 6

20 7

1.5 1.0

1.47 0.00

1.0 1.0

1.0 1.0

7.0 1.0

7

End week

72

1.5

1.52

1.0

1.0

7.0

8 0

Follow-up week Baseline

76 86

1.5 1.1

1.45 0.58

1.0 1.0

1.0 1.0

7.0 6.0

1

Week 1

64

1.4

1.50

1.0

1.0

7.0

2 3 4 5 6

Week 2 Week 3 Week 4 Week 5 Week 6

53 48 33 20 7

1.2 1.0 1.2 1.5 1.0

0.99 0.29 0.87 1.47 0.00

1.0 1.0 1.0 1.0 1.0

1.0 1.0 1.0 1.0 1.0

7.0 3.0 6.0 7.0 1.0

7 8 0 1 2

End week Follow-up week Baseline Week 1 Week 2

72 75 86 64 53

1.4 1.4 7.0 6.6 6.8

1.41 1.40 0.43 1.34 0.70

1.0 1.0 7.0 7.0 7.0

1.0 1.0 3.0 1.0 3.0

7.0 7.0 7.0 7.0 7.0

3 4 5 6 7 8 0 1

Week 3 Week 4 Week 5 Week 6 End week Follow-up week Baseline Week 1

48 33 21 7 73 76 86 64

6.9 6.8 6.8 7.0 6.6 6.6 1.0 1.4

0.87 0.88 0.51 0.00 1.38 1.25 0.34 1.24

7.0 7.0 7.0 7.0 7.0 7.0 1.0 1.0

1.0 2.0 5.0 7.0 1.0 1.0 1.0 1.0

7.0 7.0 7.0 7.0 7.0 7.0 4.0 7.0

2 3 4 5 6 7 8 0 1

Week 2 Week 3 Week 4 Week 5 Week 6 End week Follow-up week Baseline Week 1

53 48 33 21 7 73 76 86 64

1.2 1.1 1.2 1.6 1.0 1.4 1.4 1.1 1.5

0.88 0.87 0.88 1.56 0.00 1.37 1.38 0.39 1.37

1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0

1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0

6.0 7.0 6.0 6.0 1.0 7.0 7.0 4.0 7.0

Support Care Cancer (2015) 23:2589–2598

2595

Table 2 (continued)

RINV Q18

Visit week

Visit

Number, N

Mean

SD

Median

Min

Max

2 3

Week 2 Week 3

53 48

1.2 1.1

0.79 0.87

1.0 1.0

1.0 1.0

5.0 7.0

4

Week 4

33

1.2

1.05

1.0

1.0

7.0

5 6

Week 5 Week 6

21 7

1.7 1.0

1.82 0.00

1.0 1.0

1.0 1.0

7.0 1.0

7

End week

73

1.5

1.47

1.0

1.0

7.0

8 0

Follow-up week Baseline

76 86

1.4 7.0

1.35 0.43

1.0 7.0

1.0 3.0

7.0 7.0

1 2

Week 1 Week 2

64 53

6.5 6.8

1.49 0.91

7.0 7.0

1.0 2.0

7.0 7.0

3

Week 3

48

6.9

0.87

7.0

1.0

7.0

4 5

Week 4 Week 5

33 21

7.0 6.5

0.00 1.40

7.0 7.0

7.0 1.0

7.0 7.0

6 7 8

Week 6 End week Follow-up week

7 72 76

7.0 6.7 6.7

0.00 1.06 1.14

7.0 7.0 7.0

7.0 1.0 1.0

7.0 7.0 7.0

Robust relationship and correlation among variables can be seen from the biplot graph. At baseline, the two principal components can be easily distinguished (Fig. 1a). Questions 3 and 6 are also distinctly distanced from the other questions

of component 2. This indicates that nausea affects a patient’s ability to make meals and do tasks within the home similarly to a patient’s willingness to spend time with family and friends. Weekly symptom clusters

Table 3 Factor loadings and final communality from the PCA of FLIE 18-items FLIE items

Component 1

Component 2

Final communality

Q1 Q2 Q3 Q4 Q5

0.32 0.59 −0.12 0.28 0.53

0.90 0.76 0.87 0.90 0.78

0.908 0.918 0.766 0.882 0.895

Q6

−0.21

0.82

0.716

Q7 Q8 Q9 Q10 Q11 Q12 Q13 Q14 Q15 Q16 Q17 Q18 % of variance Cronbach’s alpha

0.48 0.50 0.32 0.89 0.97 0.97 0.96 0.96 0.97 0.97 0.97 0.97 69 0.962

0.80 0.82 0.79 0.38 0.21 0.21 0.16 0.16 0.21 0.17 0.21 0.21 21 0.963

0.865 0.927 0.734 0.937 0.975 0.975 0.940 0.940 0.975 0.964 0.975 0.975

The italicized entries are used to indicate which principal component each question was found to fall under.

The two components consisting of BQ1–Q9^ and BQ10–Q18^ were derived in all weeks except for week 5. At week 5, the final communality shows that all of the variables were well accounted for by one component with estimates ranging from 0.21 for Q3 to 0.96 for Q15 and Q18. The internal reliabilities for every analysis ranged between 0.99 and 0.96, which indicates excellent internal consistency. Similarly at baseline, biplots indicate that the two symptom clusters were evident at each week, with the exception of the first week of treatment. At week 1, the biplot of principal components shows that the symptom clusters were not clearly defined. This suggests that all items were experienced similarly regardless of whether the item was nausea or vomiting specific (Fig. 1b). While there were only two components for each week as expected, stronger correlations were seen between questions 3 and 6 at all weeks. This indicates that the relationship between how nausea affects household activities and personal relationships has the strongest correlation (Fig. 1c). No conclusive differences are seen between vomiting-specific items on the FLIE.

Discussion Two symptom clusters are established in patients with gastrointestinal cancers using the FLIE 18-item questionnaire. The

2596

Support Care Cancer (2015) 23:2589–2598

Fig. 1 a Biplot of FLIE 18-items at baseline. b Biplot of FLIE 18-items at week 1. c Biplot of FLIE 18-items at week 4

first component includes all vomiting items, and the second component includes all nausea items. There is dynamic change with time; however, the presence of these symptom clusters still remains, as indicated by the high internal reliability. Strong correlations are seen between questions 3 and 6 at all weeks. This indicates that nausea’s effect on one’s ability to make meals and do tasks within the home and willingness to spend time with family and friends are linked. The causes of RINV and its effects on QOL are not fully understood. It is for this reason that current antiemetic guidelines from ASCO categorize patients undergoing radiotherapy as an understudied population and recognizes the need to improve nausea and vomiting symptom control [6]. Prospective data have shown that the incidence and severity of RINV are suspected to result from a combination of radiotherapy-related factors and patient-related factors [14, 15]. These include the anatomic site of radiation, field size (>400 cm2), fractionation schedule, total dose, volume of small bowel radiated, age, gender, tumor stage, and concurrent or recent chemotherapy. At present, only Bsite of radiation^ has been incorporated into the ASCO guidelines. RINV’s effect on QOL has been poorly defined. In previous RINV studies, the effect of symptoms on QOL during and following completion of radiotherapy has only been explored to a limited extent [16, 17]. In addition, no instrument has been identified for use in gastrointestinal cancer patients receiving radiotherapy. In this study, the FLIE 18-item instrument was used as it was validated with a similar patient demographic in mind. While it has not been validated for RINV, the tools to evaluate CINV could be applied to agents of any emetogenic potential [18]. The high internal consistency of clusters throughout all weeks of the study indicates that the FLIE is a useful QOL instrument in determining the effects of RINV on patients’ QOL. Further symptom clusters could be determined by eliminating all FLIE QOL instruments with a response of 1 in Q1 or Q10. However, this would have limited the power of any findings. Despite many patients responding with no nausea or vomiting, which would obscure further correlations, a strong correlation on items 3 and 6 is still evident. For future trials, it

would be valuable to have a modified version of the FLIE to encompass a week-long period instead of 3 days. As compared to many chemotherapy agents that typically put patients at greatest risk of symptoms for a limited number of days after administration, the risk posed by radiotherapy treatments is thought to endure for a longer duration. This would ensure that all events and their relationship to patients’ QOL are captured. It is important to note that the FLIE is a questionnaire that specifically assesses QOL only as it pertains to nausea and vomiting. Therefore, it should be used in conjunction with an instrument that provides a general view of a patient’s health-related QOL. In the present study, the EORTC QLQC30 was used in addition to the 18-item FLIE. All patients treated with gastrointestinal radiation were included regardless of chemotherapy. Eighty-four percent of patients received radiotherapy that would be categorized as moderately emetogenic. In comparison, among the patients who received concurrent chemotherapy, 94 % were treated with an agent of low emetogenic potential (Table 1). Therefore, the risk of RINV is of greater concern than CINV in this setting. This study is primarily limited by the heterogeneity of the patient population and the low sample size. As the present study retrospectively explores the use of the FLIE, the study question also did not determine the sample size. In conclusion, the high internal consistency at all timepoints indicates that the FLIE QOL instrument is useful for the RINV population. However, this study may be limited by its retrospective nature and heterogeneity of treatment regimens. Future trials should also seek to explore the use of QOL measurements prospectively amongst a more homogeneously defined population.

Acknowledgments We thank the generous support of Bratty Family Fund, Michael and Karyn Goldstein Cancer Research Fund, Pulenzas Cancer Research Fund, Joseph and Silvana Melara Cancer Research Fund, and Ofelia Cancer Research Fund. Conflict of interest None.

Support Care Cancer (2015) 23:2589–2598

2597

Appendix Table 4

Table 4

18-item FLIE QOL instrument

1. How much nausea have you had in the past 3 days? 1 None

2

3

4

5

6

7 A great deal

6

7 A great deal

5

6

7 Not at all

5

6

2. Has nausea affected your ability to maintain usual recreation or leisure activities in the past 3 days? 1 None

2

3

4

5

3. Has nausea affected your ability to make a meal or do minor household repairs during the past 3 days? 1 A great deal

2

3

4

4. How much has nausea affected your ability to enjoy a meal in the past 3 days? 1 2 3 4

Not at all 5. How much has nausea affected your ability to enjoy liquid refreshment in the past 3 days? 1 2 3 4 5

7 A great deal

6

7

Not at all 6. How much has nausea affected your willingness to see and spend time with family and friends in the past 3 days?

A great deal

1 2 3 A great deal 7. Has nausea affected your daily functioning in the past 3 days? 1 2 3

7 Not at all

4

5

6

4

5

6

Not at all 8. Rate the degree to which your nausea has imposed a hardship on you (personally) in the past 3 days. 1 Not at all

2

3

4

5

7 A great deal

6

9. Rate the degree to which your nausea has imposed a hardship on those closest to you in the past 3 days. 1 2 3 4 5 6 Not at all 10. How much vomiting have you had in the past 3 days? 1 2 3 4 5 6 None 11. Has vomiting affected your ability to maintain usual recreation or leisure activities in the past 3 days? 1 2 3 4 5 6 A great deal 12. Has vomiting affected your ability to complete your usual household tasks during the past 3 days? 1 2 3 4 5 6 Not at all 13. How much has vomiting affected your ability to enjoy a meal in the past 3 days? 1 2 3 4 5 6 Not at all 14. How much has vomiting affected your ability to enjoy liquid refreshment in the past 3 days? 1 2 3 4 5 6 Not at all 15. How much has vomiting affected your willingness to see and spend time with friends in the past 3 days? 1 2 3 4 5 6 A great deal

7 A great deal 7 A great deal 7 A great deal 7 Not at all 7 A great deal 7 A great deal 7 A great deal 7 Not at all

2598

Support Care Cancer (2015) 23:2589–2598

Table 4 (continued) 16. Has vomiting affected your daily functioning in the past 3days? 1 2 3 4 5 6 Not at all 17. Rate the degree to which your vomiting has imposed a hardship on you (personally) in the past 3days. 1 2 3 4 5 6 Not at all 18. Rate the degree to which your vomiting has imposed a hardship on those closest to you in the past 3days. 1 2 3 4 5 6 A great deal

References 11. 1. Naeim A, Dy SM, Lorenz KA, Sanati H, Walling A, Asch SM (2008) Evidence-based recommendations for cancer nausea and vomiting. J Clin Oncol 26(23):3903–3910 2. Vainio A, Auvinen A (1996) Prevalence of symptoms among patients with advanced cancer: an international collaborative study. Symptom Prevalence Group. J Pain Symptom Manage 12(1):3–10 3. Enblom A, Bergius Axelsson B, Steineck G, Hammar M, Borjeson S (2009) One third of patients with radiotherapy-induced nausea consider their antiemetic treatment insufficient. Support Care Cancer 17(1):23–32 4. Feyer PC, Maranzano E, Molassiotis A, Roila F, Clark-Snow RA, Jordan K et al (2011) Radiotherapy-induced nausea and vomiting (RINV): MASCC/ESMO guideline for antiemetics in radiotherapy: update 2009. Support Care Cancer 19(Suppl 1):S5–S14 5. Dennis K, Maranzano E, De Angelis C, Holden L, Wong S, Chow E (2011) Radiotherapy-induced nausea and vomiting. Expert Rev Pharmacoecon Outcomes Res 11(6):685–692 6. Basch E, Prestrud AA, Hesketh PJ, Kris MG, Feyer PC, Somerfield MR et al (2011) Antiemetics: American Society of Clinical Oncology clinical practice guideline update. J Clin Oncol 29(31):4189–4198 7. Lindley CM, Hirsch JD, O'Neill CV, Transau MC, Gilbert CS, Osterhaus JT (1992) Quality of life consequences of chemotherapyinduced emesis. Qual Life Res 1(5):331–340 8. Clavel M, Soukop M, Greenstreet YL (1993) Improved control of emesis and quality of life with ondansetron in breast cancer. Oncology 50(3):180–185 9. Farley PA, Dempsey CL, Shillington AA, Kulis-Robitaille C, Colgan K, Bernstein G (1997) Patients' self-reported functional status after granisetron or ondansetron therapy to prevent chemotherapy-induced nausea and vomiting at six cancer centers. Am J Health Syst Pharm 54(21):2478–2482 10. Martin AR, Pearson JD, Cai B, Elmer M, Horgan K, Lindley C (2003) Assessing the impact of chemotherapy-induced nausea and vomiting on patients' daily lives: a modified version of the Functional

12.

13.

14.

15.

16.

17.

18.

7 A great deal 7 A great deal 7 Not at all

Living Index—Emesis (FLIE) with 5-day recall. Support Care Cancer 11(8):522–527 Martin AR, Carides AD, Pearson JD, Horgan K, Elmer M, Schmidt C et al (2003) Functional relevance of antiemetic control. Experience using the FLIE questionnaire in a randomised study of the NK-1 antagonist aprepitant. Eur J Cancer 39(10):1395–1401 Dodd MJ, Miaskowski C, Paul SM (2001) Symptom clusters and their effect on the functional status of patients with cancer. Oncol Nurs Forum 28(3):465–470 Poon M, Dennis K, DeAngelis C, Chung H, Stinson J, Zhang L et al (2014) A prospective study of gastrointestinal radiation therapyinduced nausea and vomiting. Support Care Cancer 22(6):1493– 1507 Roila F, Herrstedt J, Gralla RJ, Tonato M (2011) Prevention of chemotherapy- and radiotherapy-induced nausea and vomiting: guideline update and results of the Perugia consensus conference. Support Care Cancer 19(Suppl 1):S63–S65 Maranzano E, De Angelis V, Pergolizzi S, Lupattelli M, Frata P, Spagnesi S et al (2010) A prospective observational trial on emesis in radiotherapy: analysis of 1020 patients recruited in 45 Italian radiation oncology centres. Radiother Oncol 94(1): 36–41 Kirkbride P, Bezjak A, Pater J, Zee B, Palmer MJ, Wong R et al (2000) Dexamethasone for the prophylaxis of radiation-induced emesis: a National Cancer Institute of Canada Clinical Trials Group phase III study. J Clin Oncol 18(9):1960–1966 National Cancer Institute of Canada Clinical Trials Group (SC19), Wong RK, Paul N, Ding K, Whitehead M, Brundage M et al (2006) 5-Hydroxytryptamine-3 receptor antagonist with or without shortcourse dexamethasone in the prophylaxis of radiation induced emesis: a placebo-controlled randomized trial of the National Cancer Institute of Canada Clinical Trials Group (SC19). J Clin Oncol 24(21):3458–3464 Decker GM, DeMeyer ES, Kisko DL (2006) Measuring the maintenance of daily life activities using the functional living index—emesis (FLIE) in patients receiving moderately emetogenic chemotherapy. J Support Oncol 4(1):35, -41, 52