Cruickshanks KJ, Fryback DG, Nondahl DM, Robinson N, Keesey U, Dalton DS, Robertson DM, Chandra SR, Mieler WF, Zakov ZN, Custer PL, Del Priore LV, Albert DM. Treatment Choice and Quality of Life in Patients With Choroidal Melanoma. Arch Ophthalmol. 1999;117(4):461-467. doi:10.1001/archopht.117.4.461
To determine if quality of life differs between patients with choroidal melanoma treated with enucleation and those treated with radiation therapy.
Materials and Methods
Patients treated for choroidal melanoma at 5 Midwest centers were asked to participate. There were 65 participants treated with enucleation and 82 treated with radiation therapy. Quality of life was assessed using the Medical Outcome Study Short Form 36 and the National Eye Institute Visual Function Questionnaire and by the Time-Tradeoff interview method.
The average length of follow-up was 4.9 years for the group treated with radiation therapy and 6.3 years for the group treated with enucleation (P=.05). After adjusting for age, sex, years of follow-up, and the number of chronic conditions, there were few differences in any of the quality-of-life measures by treatment status. Participants in the group treated with radiation therapy were more likely to have higher (better) scores on the Vitality and Mental Component subscales of the Medical Outcome Study Short Form 36 than participants treated with enucleation. There were no differences on the National Eye Institute Visual Function Questionnaire or the Time-Tradeoff measures of quality of life.
Choice of treatment for choroidal melanoma does not seem to be associated with large differences in quality of life in long-term follow-up.
INTRAOCULAR MELANOMAS are the most common primary ocular malignant neoplasm in whites. Melanomas of the choroid and ciliary body occur in approximately 6 to 7 cases per million in the United States.1- 3 In persons older than 20 years, melanoma was the reported diagnosis for 80% of all primary ocular cancers.3 The mortality rate from malignant melanoma of the choroid depends on a variety of factors, but rates have been reported as 35% five years after enucleation and 50% ten years after enucleation.4
In the late 19th century, enucleation became the standard and almost universally accepted treatment for all choroidal and ciliary body melanomas.5 A disturbing feature of choroidal melanoma treatment is that the patient mortality rate has not decreased in the past century despite the introduction of more precise diagnostic tests allowing earlier diagnosis and improvements in enucleation techniques.6 This led Zimmerman et al7,8 to carry out a reappraisal of survival data for patients with ciliary body and choroidal melanomas. These investigators concluded that (1) the mortality rate before enucleation is low (estimated at 1% per year); and (2) the mortality rate rises abruptly following enucleation, reaches a peak of about 8% during the second year after enucleation, and then decreases monotonically.7,8 They further concluded from their studies that about two thirds of the fatalities following enucleation can be attributed to the dissemination of tumor emboli during surgery. These assumptions have been challenged by other investigators.9- 11 Subsequent examination of available data at several eye tumor treatment centers left the question unsettled, as many of these were small studies of short duration and without randomization.12,13
This controversy spurred the development of alternative methods of treatment for intraocular melanoma, particularly radiation therapy. To resolve this issue, the National Eye Institute, National Institutes of Health, Bethesda, Md, undertook a multicenter, randomized clinical trial, the Collaborative Ocular Melanoma Study (COMS),14 to compare survival rates between patients treated with enucleation and those treated with radiation therapy for medium-sized tumors. The COMS involves 42 major centers throughout the United States and Canada and is now in its 12th year.
Patients with choroidal melanoma and their ophthalmologists are confronted with a difficult dilemma regarding treatment choice when the decision is not made by randomization. On one hand, patients often fear that enucleation will be disfiguring despite assurances from their ophthalmologist of a good cosmetic result. Enucleation also raises fears of eventual poor vision or blindness in the fellow eye, even when this eye is healthy. Radiation therapy, on the other hand, causes concerns about residual viable tumor, compromised vision, ocular discomfort, and chronic ocular irritation. The cost difference between the 2 methods of treatment is considerable, with the cost of radiation therapy approaching twice that of enucleation (Michael Drummond, PhD, and Linda Davies, MSc, written communication, 1990). Clearly, relative quality of life and emotional outcome, including physical function, role function, social and emotional function, and level of health problems must be considered in the treatment decision.15 However, little information regarding quality of life of choroidal melanoma patients is otherwise available.16- 18 Although a quality of life assessment has been incorporated into COMS,19 data from this study may not be forthcoming for a considerable time. For this reason, we carried out a retrospective study on patients unconnected with the COMS to determine if quality of life differed between patients treated for choroidal melanoma with enucleation and those treated with radiation therapy.
Patients who had been treated for choroidal melanoma with enucleation or radiation therapy at 1 of 5 clinics and who were not enrolled in the COMS were identified from 5 large ocular oncology practices in the Midwest: the University of Wisconsin Medical School (Madison), Mayo Clinic (Rochester, Minn), the Medical College of Wisconsin (Milwaukee), Retina Associates of Cleveland (Beachwood, Ohio), and Washington University (St Louis, Mo). The protocol was reviewed and approved by each institution's human studies committee. Patients were invited to participate by a letter from their physician. To enroll, they were asked to make a collect call to a number at the University of Wisconsin Medical School and to provide their names and mailing addresses. Patients were assigned a code number and all data were identified only by this number.
Of the 370 patients contacted by their physicians, 147 (40%) participated in this study. There were no statistically significant differences between responders and nonresponders in the mean year of treatment or age at treatment. Nonresponders were slightly more likely than responders to have been treated with enucleation (51.4% vs 42.6%, P=.09).
Participants were asked to complete a mailed questionnaire, which consisted of the Medical Outcome Study (MOS) 36-Item Short Form Health Survey (SF-36)20,21 and the National Eye Institute Visual Function Questionnaire (NEI-VFQ; 51 items).22 Half of the patients were asked to complete the NEI-VFQ first and half were asked to complete the MOS SF-36 first.
On receipt of the questionnaire, the data were reviewed for completeness. Respondents were sent a set of props for the Time-Tradeoff interview (TTO).23,24 This consisted of 5 colored index cards labeled A through F with the description of a "trade" printed on them. These were sent in a sealed envelope with instructions not to open them until the interview. This questionnaire has been used in other studies to assess the relative value of various health states by asking a series of questions about the number of years (out of 10 years) a person would be willing to trade in a particular health state to have a better health state. The interviewer was not aware of the participant's treatment modality.25- 27
All data were double-entered and discrepancies were adjudicated and resolved. The MOS SF-36 questionnaire was scored in standard fashion, resulting in eight 0 to 100 subscales: General Health, Physical Functioning, Social Functioning, Mental Health, Physical Role, Emotional Role, Bodily Pain Index, and Vitality. In addition to the 8 subscales, 2 additional subscales were computed: the standardized Physical Component Scale and the standardized Mental Component Scale. These 8 subscales were standardized based on US population data and weighted to obtain 2 continuous scales that, in our data, ranged from 17 to 59 (Physical Component Scale) or 13 to 67 (Mental Component Scale).28 All 8 subscales are used in the scoring of both the Physical Component Scale and Mental Component Scale. The subscales contributing the strongest positive factor loadings for the Physical Component Scale were Physical Functioning, Physical Role, Bodily Pain Index, and General Health; and for the Mental Component Scale were Mental Health, Emotional Role, Social Functioning, and Vitality. In addition, the estimated Quality of Well-being score was computed using a published regression model.29
The NEI-VFQ was scored in standard fashion, resulting in 10 0 to 100 subscales: General Health, General Vision, Visual Pain, Near Vision, Distance Vision, Vision Social Function, Vision Mental Health, Vision Role, Vision Dependency, and Driving. In addition, a 25-item, global 0 to 100 scale was developed.
The introduction to the NEI-VFQ asked participants, "Has a doctor ever told you that you have any of the following conditions or problems?" The conditions were arthritis/rheumatism, cancer (other than skin cancer), major paralysis or neurologic problems (such as stroke/epilepsy, multiple sclerosis, muscular dystrophy), cardiac pacemaker, amputation of an arm or leg, heart failure or enlarged heart, heart attack or angina, asthma or other serious lung problems (such as chronic bronchitis or emphysema), back problems (including disk or spine problems), ulcer (duodenal, stomach, or peptic), chronic inflamed bowel/enteritis/colitis, kidney or liver disease, diabetes, high blood pressure, deafness or trouble hearing, or other major health problems. Each participant was assigned a score (0-16) based on the number of chronic conditions reported. This score was used in selected subsequent analyses to adjust for health status, as described below.
For the TTO, patients equated 2 prospects, a fixed prospect of being in a worse health state for 10 years vs 1 in an ideal health state for less than 10 years with the amount of time adjusted until indifference was reached (eg, 8 years). The TTO score was the percentage of 10 years assigned to the ideal health state at the point of indifference (eg, 80%). Three health states were explored as to how they compared with excellent vision and health: (1) current vision/health, (2) blind/current health, and (3) perfect vision/current health otherwise. In addition, the quantity (1-2)/(3-2) can be interpreted as the relative value given to current visual health, with 0.0 indicating blindness and 1.0 indicating perfect vision. The dichotomous outcome, "willing to trade any time at all," was analyzed for each of the 3 health states. Participants were also asked to rank and then rate the 3 health states, with values ranging from 0 (death) to 100 (excellent health).
Because most quality of life outcomes were assessed on a continuous scale, t tests of least squares means from analysis of covariance models were used to evaluate treatment differences. Two sets of analyses were conducted: one that adjusted for age and sex, and one that adjusted for age, sex, follow-up time, and number of chronic conditions. One exception was the TTO evaluation of willingness to trade any time. Logistic regression models were used to assess this dichotomous outcome. Data analyses were conducted using statistical computer software (SAS Language, Version 6.09; SAS Institute Inc, Cary, NC).30 Analyses using nonparametric methods yielded similar results (data not shown).
The characteristics of participants by treatment group are given in Table 1. There were no differences in age, sex, number of chronic health conditions, or self-rated general health. Participants treated with radiation therapy had fewer years of follow-up care than the enucleated group (4.9 vs 6.3 years, P<.05). The distribution of treatment type differed by clinic site.
There were few differences in these measures of quality of life between those treated with radiation therapy and those treated with enucleation (Table 2). After adjusting for the potential confounding effects of age, sex, length of follow-up, and number of chronic conditions, people treated with radiation therapy were significantly more likely to score slightly higher (better) on the Vitality domain and standardized Mental Component Scale of the MOS SF-36 (P<.05). The Vitality scale is derived from 5 questions, "How much of the time during the past 4 weeks did you feel full of pep," "did you feel worn out," "did you have a lot of energy," and "did you feel tired," which can be rated from 1 (none of the time) to 5 (all of the time). There were no statistically significant differences between treatment groups in the estimated Quality of Well-being score, the NEI-VFQ, or the TTO questionnaire.
There were insufficient numbers of participants in each treatment group per clinic to control for clinic site. To confirm that these differences in quality of life were not simply marking patient differences between sites, we analyzed the data for the Mayo Clinic (the only site with similar numbers of patients in each treatment group). As given in Table 3, follow-up time was longer for those treated with enucleation than for those treated with radiation therapy. Results from this subgroup analysis were consistent with those for the whole study, although the differences in Vitality and the Mental Component Scale on the MOS SF-36 were no longer statistically significant (Table 4).
When reviewing results of the TTO interview, there were 43 participants who gave responses that lacked internal consistency (for example, inconsistencies between the 3 methods [ranking, feeling thermometer, and TTO scores] or participants who did not prefer perfect vision to current vision or blindness). To evaluate the potential effect of these reports on the results, we excluded these participants and reanalyzed the data. Results of these analyses again demonstrated a significant difference in the Vitality Scale on the MOS SF-36 with patients treated with radiation therapy having higher scores (data not shown).
For more than a century, enucleation has been the standard treatment for choroidal melanoma.5 Because of the concern for the potential for tumor spread resulting from manipulation of the eye during enucleation7 combined with concerns regarding loss of vision and poor cosmetic result, radiation therapy has become a popular alternative. The risks and benefits of radiation therapy compared with enucleation have yet to be established.5 When patients make a treatment decision regarding choroidal melanoma, major factors are fear of death due to cancer and concerns regarding prognosis for vision in the affected eye.19 Economic factors may become a consideration as well. It is our impression that ophthalmologists assume that if the treatments were equally effective, the results would be better with radiation therapy, since this form of treatment potentially allows the retention of some vision and avoids the perception of disfigurement.
This retrospective cohort study was performed to determine whether patients treated with enucleation for choroidal melanoma have a relatively better quality of life than those treated with radiation therapy. Although accrual in the COMS was recently completed, it will be several years before outcome data regarding survival, vision retention, or quality of life are available. The COMS captures approximately one half of the new cases of ocular melanoma occurring during the enrollment period in the United States. The remaining patients are treated in a nonrandomized manner determined by the patient's desire and the opinions and clinical experience of the physicians.
An additional consideration that may become increasingly important is cost. No studies have been published on the differences in cost between enucleation and radiation therapy in choroidal melanoma. Two health economists from England, Michael Drummond, PhD, and Linda Davies, MSc, began an economic evaluation of patients in the COMS at the request of the National Eye Institute. Preliminary data indicate that the average cost of enucleation is $7300 and the average cost of radiation therapy $13,500 (Michael Drummond, and Linda Davies, written communication, 1990).
The 3 most important determinants of the "success" of treatment for choroidal melanoma are survival, vision, and quality of life. In comparing enucleation with radiation therapy in terms of survival, numerous case series suggest little difference.5 Initial mortality data from the COMS, a randomized comparison of the 2 methods, found no difference in 5-year mortality for patients with large choroidal melanoma.31 If the final data from COMS, a randomized comparison of the 2 methods, show similar results, vision and quality of life will then be the important outcome factors in determining the procedure of choice.
Most patients who survive choroidal melanoma retain satisfactory vision for many years after initial treatment, regardless of whether they are treated with enucleation or radiation therapy.32 Edwards and Schachat17 carried out a "functional outcome assessment" of 71 enucleated patients. Patients were interviewed with regard to working, driving, reading, and television viewing. Eighty-seven percent reported no change in their ability to perform these vision-dependent tasks. Fifteen years after enucleation, 90% retained the ability to drive and 96% retained the ability to read.
For the purpose of this study, we accepted the definition of quality of life to be an aggregate representation of disability, discomfort, and distress resulting from illness and/or medical and surgical action.33 In assessing quality of life in this study, 3 instruments were used: (1) the MOS SF-36 Health Survey20,21; (2) the NEI-VFQ22; and (3) the TTO method of cost-utility analysis.23,24 The MOS SF-36 was designed for use in clinical practice and research, health policy evaluations, and general population surveys. The MOS SF-36 contains 1 multi-item scale that assesses 8 health concepts: (1) limitations in physical activities because of health problems (Physical Functioning); (2) limitations in social activities because of physical or emotional problems (Social Functioning); (3) limitations in usual activities because of physical health problems (Physical Role); (4) Bodily Pain; (5) Mental Health (psychological distress and well-being); (6) limitations in usual activities because of emotional problems (Emotional Role); (7) Vitality (energy and fatigue); and (8) General Health perceptions.
The NEI-VFQ is a self-administered version of a questionnaire developed in conjunction with the RAND Corporation, Santa Monica, Calif, and has been field-tested across persons with 1 of 5 eye diseases in the United States.22 When direct measurement by testing in the office or by mail is not feasible, this questionnaire offers a proxy for measured visual function. Preliminary information indicates that this test responds to visual function.22 This relationship complements the MOS SF-36, which has no formal visual function assessment.
The TTO method of quality of life assessment is a recognized method of obtaining utility values for health states.25- 27 It was developed specifically for use in health care by Torrance et al.34
These assessment tools operationalize the concept of quality of life differently. The MOS SF-36 and the NEI-VFQ are questionnaires that ask subjects to rate the degree of limitation or function with respect to various physical, mental, emotional, and visual tasks. The TTO method addresses preference or utility of the subject for their overall state of health. In a sense, the 2 questionnaires define quality of life as "what you can do and how you feel," and the TTO defines it as "how much you like what you can do and how you feel." The patients treated with enucleation seemed to have somewhat less energy and more fatigue than the patients treated with radiation therapy on the MOS SF-36 Vitality scale. Ware et al21 report that a 5-point difference between groups is clinically and socially relevant, which suggests that the lower Vitality score for patients treated with enucleation may be clinically significant. However, the overall pattern shows minimal differences between the 2 treatment groups with both operational definitions of quality of life considered in the aggregate.
There are, however, potential shortcomings inherent in this retrospective study as well as in the specific instruments used. Overall response to the survey was low. While participants did not differ from nonparticipants in type of treatment, age at treatment, or follow-up time, the possibility of selection bias remains. While this is the largest survey of quality of life in choroidal melanoma patients to date, because of the low response rate and the exclusion of COMS participants, results from this study cannot be extrapolated to all patient populations.
In these analyses, vision status was not included as a confounding variable, as only self-reported visual function, and not physician-measured visual acuity, was available. Thus, the contributions of impaired vision to quality of life were considered part of the potential effect of treatment type. Any differences in vision status between treatment groups would have contributed to finding a treatment-related effect on quality of life. Since there were few differences in any of the outcome measures, these data suggest that there is little difference by treatment type in long-term quality of life among choroidal melanoma patients. In comparison with published normative data for people of similar age or with other chronic conditions (for example, acute myocardial infarction or hypertension), mean scores for participants in each treatment group were as high or higher (better) on each subscale of the MOS SF-36 (Table 5).21 These qualitative comparisons suggest that as a group, the overall quality of life among choroidal melanoma patients is comparable to others in their age range.
Although the number of chronic conditions affecting the patient was included as a covariate in multivariate analyses, there may remain some residual confounding, as this may be an inadequate marker of comorbidity. However, significant bias is not likely to be introduced, as it is unlikely that there would be large differences in morbidity status between treatment groups.
There also seemed to be some inconsistency within individuals in the TTO responses that point to a limitation in the methods used in this study, since the methods have been used successfully elsewhere in very ill and/or very elderly patients.26,35 We suspect that these problems reflect the difficulties of conducting the questionnaire via the telephone and/or applying the questionnaire to this particular disorder (ie, choroidal melanoma). Nonetheless, results excluding these participants were similar, showing little difference in quality of life between treatment groups.
Our study does not confirm the report by Trunc et al18 comparing quality of life outcomes for different uveal melanoma treatment alternatives. In their study, almost 100 patients were stratified according to treatment and visual acuity. The study assessed work performance, difficulty in night driving, cosmetic appearance of prosthesis and eyelash loss, among other factors. The authors concluded that radiation therapy with preserved vision maintains the best quality of life in uveal melanoma patients. The difference in conclusions may be due to the different instruments used in the 2 studies, different biases involved in determining which patients were referred for each modality of treatment (ie, whether patients with fewer comorbidities were treated with radiation therapy), or actual differences in the radiation therapy or enucleation techniques used by the 2 groups. The quality of life assessment being carried out by COMS as part of a prospective randomized clinical trial should provide more definitive data.
Ultimately, the recommendation that ophthalmologists make to their patients with choroidal melanoma will be based on the data collected for survival, retention of vision, and quality of life. Economic factors may be a consideration as well. This study indicates that with regard to quality of life as determined by the 3 instruments used, little difference existed between the patients treated with enucleation and those treated with radiation therapy that we studied.
Accepted for publication September 15, 1998.
This study was supported by Research to Prevent Blindness Inc, New York, NY, and the Wisconsin Lions Eye Research Fund, Madison.
Reprints: Karen J. Cruickshanks, PhD, Department of Ophthalmology and Visual Sciences, University of Wisconsin, 610 N Walnut St, 460 WARF, Madison, WI 53705-2397.