Uveal Melanoma: Refusal of Treatment

Refusal of treatment (enucleation) for uveal melanoma has been well recorded in the landmark treatise by Fuchs [1]. Zimmerman and McLean [2] cataloged 29 cases from the USA and reviewed cases from Europe in their 1979 manuscript reporting metastatic disease in untreated uveal melanoma. Among the European cases, all 10 cases from Germany that refused enucleation were dead within 10 years [3]. Four of the 5 cases from Finland died due to metastasis within 2–15 years [4]. By reading through brief clinical histories of US cases, only 2 (cases 5 and 6) can be identified as accounts of patients that refused treatment (enucleation) [2]. Other included cases were either misdiagnosed on prior examinations or diagnosed on autopsy that was performed following death due to metastatic disease. Both patients survived 7 years after initial diagnosis when they presented with ocular pain necessitating enucleation. Both had developed extraocular extension and died of metastases.

Review of the more recent literature indicates that refusal of ocular treatment in uveal melanoma is exceptionally uncommon [5]. The only reported data are from the COMS (Collaborative Ocular Melanoma Study) natural history study with an estimated rate of 0.9% (77 cases out of a total of 8,712 evaluated cases) [5, 6]. A statistically significant difference in the all-cause mortality and metastasis-free survival (MFS) in those who refused treatment from those who underwent treatment could not be established due to the small number of cases (45). As only basic descriptive information on age, sex, and race was obtained, the COMS natural history study lacked sociodemographic information to identify the factors for refusing tumor treatment and the conversion factors for eventually accepting treatment [5].

Refusal of treatment in cancer is multifaceted. Reasons for refusal of care may include financial reasons, sociodemographic factors, limitation of transportation, mental health, physical restriction, fear of losing vision, comorbidities affecting health resulting in a state of overwhelm, denial of disease, and capacity to understand the disease, prognosis, and treatment [7-9]. We are unable to find any study evaluating reasons for refusal of ocular treatment in patients with uveal melanoma. The aim of this study was to evaluate the factors for refusing recommended ocular treatment in patients with uveal melanoma. Secondary analysis for ocular survival, MFS, and overall survival (OS) was also performed.

We conducted a retrospective study of 9 patients diagnosed with primary uveal melanoma at our institution from October 2009 to June 2018. These patients were included as they had declined primary ocular therapy including thermotherapy, plaque brachytherapy, or enucleation. Specifically, patients who underwent recommended treatment elsewhere or individuals who were observed as part of the management plan were excluded.

Our primary research aim was to discover factors for patients refusing primary recommended treatment. The secondary aim comprised reporting ocular survival, MFS, and OS. Tumor-related data included tumor location, size, COMS classification, primary recommended treatment, patient-stated reason for refusal of treatment, subsequent course, and treatment (if any). Mental health disorder as listed in existing diagnosis was retrieved. Mental health issues had been recorded as stated by the patients or their accompanying family members as part of general history taking without specific attempt to evaluate patients for mental disorders. The last patient contact and mortality data were obtained by review of the electronic medical records and review of scanned documents received from outside care providers. For patients who were deceased, the date of death was determined by review of the electronic medical record, obituaries, or notification by family.

Sociodemographic data collected included age at diagnosis, sex, ethnicity, marital status, insurance type (Medicare/military/veteran’s association, private/others, or Medicaid/none/self), distance of primary residence (estimated using Google maps to calculate miles from the patient’s primary address zip code to Cleveland Clinic’s zip code [44,195] and then divided into 3 groups: 0–100 miles, 101–200 miles, and 201 + miles), median household income based on the zip code (website www.incomebyzipcode.com), and Charlson Comorbidity Index (CCI) score (https://www.mdcalc.com/charlson-comorbidity-index-cci) which predicts 10-year survival based on comorbidities [10]. Although socioeconomic factors include education, housing, employment, and income [11, 12], we were limited only to average income based on geographic location. For comparison, we identified a control group of 107 consecutive uveal melanoma patients who underwent treatment as per our recommendations (January 2014 to June 2015) [10].

A swimmer plot was created to depict the timeline of patient outcome that included time of refusal of primary recommended treatment, subsequent treatment (if any), and status at last contact with the patient (alive/dead; melanoma metastasis present/absent). Statistical analysis was performed with the t test for unequal variances for age, median income, and CCI. The categorical values for marital status, insurance, and distance were classified as 0 or 1 for which the t test of unequal variance was performed. Analyses were performed using Excel (Microsoft® Software 2018, NY, USA).

From the chart review, 9 patients met inclusion criteria for refusing primary recommended treatment for uveal melanoma after the risks and benefits were discussed with the patient. The mean age at diagnosis was 69 (range: 35–82), with 56% male. The average time of follow-up from initial diagnosis when primary recommended treatment was refused until the last clinic follow-up was 22 months. Visual acuity at initial diagnosis was 20/30 or better in 5 cases, 20/40–20/50 in 2 cases, and CF or worse in 2 cases. The choroidal melanomas were small (n = 1 [11%]), medium (n = 5 [56%]), and large (n = 3 [33%]) in size (COMS criteria) (Table 1).

Statistical analysis was performed between the sociodemographic profile of the 9 patients who refused uveal melanoma treatment and a cohort of 107 patients who received treatment. There was no difference in mean age (61 vs. 69; p = 0.13), sex (p = 0.56), ethnicity (p = 0.39), marital status (p = 0.28), insurance type (p = 0.37), distance (p = 0.85), median income (USD 55,312 vs. USD 54,232; p = 0.42), or Charlson Comorbidity Index score (3.1 vs. 3.7; p = 0.39), in the treated verses refused treatment group (Table 2).

Patient-stated reasons for refusal of treatment included lack of family support (n = 1 [11%]), co-existing health issues (n = 2 [22%]), and desire to pursue alternative nontraditional approaches (n = 2 [22%]). Five patients did not perceive immediate therapeutic benefit, either because they had excellent vision (4, >20/40), which was likely to get worse after treatment, or had poor vision (1, light perception), which was unlikely to improve after treatment (Table 3).

One patient (11%) had pre-existing mental health diagnosis of severe anxiety, depression, and attention deficit disorder (Table 3).

At initial assessment, plaque brachytherapy was recommended for 7 patients (78%) with enucleation for other 2 patients (22%) (Table 1). Due to tumor growth, at the last follow-up, plaque brachytherapy could only be recommended for 4 patients (45%) with enucleation in the other 5 patients (55%) (Table 1). In total, 5 patients (55%) eventually accepted treatment following a mean delay of 19 months (range: 4–55 months). Four underwent enucleation due to large tumors (3 for uncontrolled pain and 1 for loss of vision), and 1 patient with a juxtapapillary macular tumor who had an initial visual acuity of 20/30 vision developed count fingers vision and re-presented and consented for plaque brachytherapy and transpupillary thermotherapy. Four patients (45%) refused treatment at the last follow-up due to stable good vision (2.22%) (Fig. 1) and stable poor vision (1.11%), and 1 patient died of unrelated causes (Table 3).

The refusal of primary therapy for an average period of 19 months (range: 4–55 months) in this cohort was associated with worse ocular survival with more than doubling of the enucleation rate from 22% (2/9) at initial evaluation to 55% (5/9) at the last visit (Table 1).

Three patients died over an average of 4.2 years (range 12–124 months), one with liver metastasis which was detected during an emergency room evaluation. Presence or absence of metastasis in other patients could not be assessed due to lack of periodic systemic surveillance. The remaining 6 patients were alive at last contact with an OS of 67%.

We retrospectively studied 9 patients (representing about 0.8% of all uveal melanoma cases) who were diagnosed with primary uveal melanoma yet declined ocular treatment. The choroidal melanoma was predominantly small or medium (67%), and globe-sparing episcleral plaque brachytherapy was recommended in the majority (7.79%) of cases (Table 1). We were intrigued as to why patients would decline treatment (sometimes repeatedly) while the majority would travel great distances to seek expert care.

Sociodemographic factors, including non-White race, male sex, increasing age, lack of health insurance, higher Charlson Comorbidity score, unmarried status, and elderly age, have been associated with increased risk of refusing recommended breast, lung, prostate, and colon cancer treatment [9, 13]. Limited sociodemographic analysis (age, sex, and ethnicity) in the COMS natural history study was not revealing [5, 14]. Even with the expanded sociodemographic profile assessed in our study, on statistical comparison of the 9 patients who refused treatment with a cohort of 107 patients who received treatment, no significant difference was observed in mean age (61 vs. 69; p = 0.13), sex (p = 0.56), ethnicity (p = 0.39), marital status (p = 0.28), insurance type (p = 0.37), distance (p = 0.85), median income (USD 55,312 vs. USD 54,232; p = 0.42), or Charlson Comorbidity Index score (3.1 vs. 3.7; p = 0.39) [10].

One patient (11%) had existing mental health diagnosis of severe anxiety, depression, and attention deficit disorder (Table 3) [15]. The patient was willing to continue follow-up (Fig. 2) but continued to refuse treatment. Perhaps others may suffer from as yet undiagnosed or subclinical mental health disorder which has been overlooked as these patients were not subjected to any formal psychiatric evaluation.

Patient-stated reasons for refusal of treatment such as lack of family support for transportation (1.11%), co-existing health issues (2.22%), and desire to pursue alternative nontraditional approaches (2.22%) became evident. Lack of transportation has been reported in the literature as an important contributory factor [15]. Another perhaps overlooked factor seems to be patients lacked perception of immediate therapeutic benefit particularly when they have good vision (4, >20/40) that is likely to get worse after treatment or if they have poor vision (1, LP) which is unlikely to improve after treatment. The refusal to accept treatment persisted even after discussion of potential risks of metastasis and death. The subsequent course of the 5 patients who eventually accepted treatment following an average delay of 19 months (range: 4–55) highlights the conversion factors: pain and loss of vision. Two patients who have maintained good vision (20/50 or better) continue to refuse treatment despite evidence of tumor growth; however, they have demonstrated willingness to return for periodic assessment. These patients clearly understood their condition, impact of melanoma diagnosis, and need for treatment, yet they choose to not proceed with treatment due to the fear of losing vision secondary to radiation and/or surgery.

The refusal of primary therapy for an average period of 19 months (range: 4–55 months) in this cohort was associated with worse ocular survival with more than doubling of the enucleation rate from 22% (2/9) at initial evaluation to 55% (5/9) at the last visit. Three people died over an average of 4.2 years (range: 12–124), one with liver metastasis which was detected during an emergency room evaluation. Our preferred outcome measure of MFS could not be assessed due to lack of periodic systemic surveillance in these patients. Similar to results of the COMS natural history study, we were unable to ascertain adverse effect on OS due to small number of patients.

Refusal of uveal melanoma treatment is multifactorial. Refusal of initial recommended treatment is associated with poor ocular survival. The small sample size did not allow for an evaluation of the impact on survival.

The research complies with the guidelines for human studies, and the research was conducted ethically in accordance with the World Medical Association Declaration of Helsinki. The study protocol was approved by the Cleveland Clinic Institutional Review Board (20-722). As this study involved data collection from the existing records, written consent of the patients was not required.

Prof. Arun D. Singh is the Editor-in-Chief of Ocular Oncology and Pathology. Arun D. Singh has reported having relevant financial activities outside the submitted work: Aura Biosciences (stock options), IsoAid LLC (consultancy), Immunocore (consultancy), Isoaid (consultancy), and Eckert and Zeigler (consultancy). Randy Christopher Bowen, Hansell Soto, Vishal Raval, and Jacquelyn M. Davanzo have no reported relevant financial activities.

This work was supported by an unrestricted departmental grant from the Research to Prevent Blindness Challenge Grant, to the Cole Eye Institute, Cleveland Clinic.

All authors contributed to the diagnosis (R.C.B. and A.D.S.) and management (R.C.B., J.M.D., and A.D.S.) of the described patients. All authors also contributed to collection of data (R.C.B. and J.M.D.) and writing and/or editing of the manuscript (R.C.B., H.S., V.R., J.M.D., and A.D.S.).

Data are available upon request.