Key Points
Question
What is the association between surgical excision of Tis and T1a melanomas of the head and neck and the health-related quality of life of a patient?
Findings
In this longitudinal cohort study involving 56 patients with a Tis or T1a melanoma of the head and neck, patient-reported health-related quality of life was worse in the perioperative period but improved to above baseline levels by 6 months and 1 year after excision of the tumor. Women and those younger than 65 years experienced worse health-related quality of life over the treatment course.
Meaning
Surgical excision of Tis and T1a melanomas of the head and neck may be associated with long-term net improvement of health-related quality of life; baseline or preoperative counseling may be tailored to groups at risk of lower health-related quality of life.
Abstract
Importance
Surgical excision is the standard-of-care treatment for Tis and T1a melanomas of the head and neck. Currently, however, the association of diagnosis and surgical treatment of these typically slowly progressive and nonfatal melanomas with a patient’s health-related quality of life (HRQoL) is unknown.
Objective
To characterize and assess HRQoL in patients with Tis and T1a head and neck melanoma, evaluate changes in HRQoL over the surgical treatment course, and identify patient characteristics associated with lower HRQoL.
Design, Setting, and Participants
This longitudinal, prospective cohort study involved patients with Tis or T1a melanoma of the head and neck who underwent staged excision at a single tertiary care center (Memorial Sloan Kettering Cancer Center, New York, New York) and were recruited from June 1, 2016, to February 28, 2017. Patients were followed up for 1 year after their surgical procedure. Participants were asked to complete 2 patient-reported outcome measure questionnaires, Skindex-16 and Skin Cancer Index (SCI), at 4 time points: baseline, perioperative (1 to 2 weeks after surgery), and 6-month and 1-year follow-up.
Main Outcomes and Measures
Scores on the Skindex-16 and SCI questionnaires.
Results
In total, 56 patients were included in the study, among whom 24 (43%) were female and 32 (57%) were male, with a mean (range) age of 67.2 (32-88) years; all patients self-identified as white. Forty-one (73%) questionnaires at perioperative, 49 (88%) at 6-month postoperative, and 41 (73%) at 1-year postoperative time points were completed. At baseline, female patients and those younger than 65 years had statistically significantly worse HRQoL on the Skindex-16 questionnaire (mean score, 14.2 [95% CI, 9.1-21.9] and 16.1 [95% CI, 9.8-26.4]) and on the SCI questionnaire (mean score, 57.2 [95% CI, 48.3-67.6] and 53.2 [95% CI, 44.1-64.3]) compared with males (mean Skindex-16 score, 7.0 [95% CI, 4.8-10.3]; mean SCI score, 73.5 [95% CI, 66.0-81.7]) and those aged 65 years or older (mean Skindex-16 score 7.1 [95% CI, 5.0-10.0]; mean SCI score, 74.3 [95% CI, 67.7-81.6]). Questions that demonstrated the worst scores at baseline were worry about skin condition (Skindex-16) and worry about future skin cancers (SCI). The emotions subscale scores on the Skindex-16 questionnaire showed the greatest improvement from baseline to 1-year follow-up levels (26.6 vs 15.3; P < .001) and so did the appearance subscale scores on the SCI questionnaire (64.0 vs 84.6; P < .001). The score difference in HRQoL by sex diminished over time, whereas the score difference by age persisted through the first year.
Conclusions and Relevance
Improvement in HRQoL at the 6-month and 1-year follow-up was associated with surgical excision in patients with early-stage head and neck melanoma, and younger and female patients experienced worse HRQoL. These results may be used in tailoring counseling for this patient population.
This study examines the preoperative and postoperative changes in quality of life as perceived and reported by patients who received surgical treatment for Tis or T1a melanoma of the head and neck.
Introduction
Tis and T1a melanomas of the head and neck have an excellent prognosis, with more than a 98% five-year disease-specific survival1; however, the association of treatment with health-related quality of life (HRQoL) is not currently known. Melanoma in sun-exposed areas of the head and neck is known for substantial subclinical extension and thus presents distinct treatment challenges. Standard excision margins are often inadequate for clearance of melanoma in these locations, resulting in large surgical defects and reconstruction.2,3,4 These considerations, as well as the common public perception of melanoma as a life-threatening cancer, may uniquely affect patient HRQoL. Health-related quality of life impairment among patients with melanoma may be largely psychological,5 as patients have reported high levels of anxiety and distress even before receiving histologic confirmation of their diagnosis.6 Scarring, particularly in conspicuous areas, must also be considered. A better understanding of patients’ HRQoL throughout the diagnosis, treatment, and recovery processes will enable clinicians to provide comprehensive, patient-centered care. Therefore, the objectives of this study were to characterize and assess HRQoL in patients with Tis or T1a melanoma of the head and neck, evaluate the long- and short-term changes in HRQoL over the course of surgical treatment, and describe the patient characteristics associated with lower HRQoL.
Methods
This longitudinal, prospective cohort study was approved by the Institutional Review Board of Memorial Sloan Kettering Cancer Center. Written patient consent was obtained from patients by Memorial Sloan Kettering Cancer Center.
Patients who underwent staged excision at Memorial Sloan Kettering Cancer Center (New York, New York) for the treatment of Tis or T1a stage7 of melanoma of the head and neck region were consecutively recruited from June 1, 2016, to February 28, 2017. The patient-reported outcome measure (PROM) instruments used were the Skindex-16, a dermatologic HRQoL questionnaire that quantifies the changes in skin conditions over the past week,8 and the Skin Cancer Index (SCI), a validated HRQoL questionnaire for patients with nonmelanoma skin cancer.9 Higher scores indicate worse HRQoL in Skindex-16 but better HRQoL in SCI. All participants completed a set of PROM instruments prior to the surgical procedure, and the questionnaires were administered again at the 1- to 2-week perioperative time point as well as at the 6-month and 1-year postoperative time points. If the patient was unable to come to the follow-up, the questionnaire was mailed to the patient. The following data were collected from the medical record of each patient: age, sex, race/ethnicity, marital status, comorbidities, history of anxiety or depression, tumor location, and biopsy result.
Descriptive statistics were used to describe the patient population and the results of the questionnaires. Geometric means and SDs were used because of the skewed nature of the Skindex-16 and SCI distributions. Unpaired, 2-tailed t tests and 1-way analysis of variance were used to assess differences in the distribution of scores between subsets of the patient population. A 2-sided P < .05 was used to indicate statistical significance. Separate random-effects regression models were used to assess the change in HRQoL measures over time. The potential interaction between patient sex and evaluation time point was assessed. Model-based predictive marginal probabilities were plotted to visualize the trajectories of scores for male and female participants. All analyses were performed using Stata, version14.2 (StataCorp LLC).
Results
Fifty-six patients completed baseline PROM instruments prior to their surgical procedure. Among these patients, 24 (43%) were female and 32 (57%) were male, with a mean (range) age of 67.2 (32-88) years; all patients self-identified as white. At baseline, female patients and those younger than 65 years had statistically significantly worse HRQoL on the Skindex-16 questionnaire (mean score, 14.2 [95% CI, 9.1-21.9] and 16.1 [95% CI, 9.8-26.4]) and on the SCI questionnaire (mean score, 57.2 [95% CI, 48.3-67.6] and 53.2 [95% CI, 44.1-64.3]) compared with males (mean Skindex-16 score, 7.0 [95% CI, 4.8-10.3]; mean SCI score, 73.5 [95% CI, 66.0-81.7]) and those aged 65 years or older (mean Skindex-16 score 7.1 [95% CI, 5.0-10.0]; mean SCI score, 74.3 [95% CI, 67.7-81.6]). (Table 1). No substantial baseline differences by other patient or clinical characteristics were observed.
Table 1. Select Patient Characteristics and Mean Baseline Skindex-16 and Skin Cancer Index (SCI) Scores .
| Variable | No. (%) | Skindex-16 Score | Skin Cancer Index Score | ||
|---|---|---|---|---|---|
| Geometric Mean (Geometric SD) | P Value | Geometric Mean (Geometric SD) | P Value | ||
| Age, y | .001c | .007c | |||
| <65 | 20 (36) | 16.1 (2.9) | 53.2 (1.5) | ||
| ≥65 | 36 (64) | 7.1 (2.8) | 74.3 (1.3) | ||
| Sex | .03c | .008c | |||
| Female | 24 (43) | 14.2 (2.8) | 57.2 (1.5) | ||
| Male | 32 (57) | 7.0 (2.9) | 73.5 (1.3) | ||
| Visibility of site | .54c | .46c | |||
| Inconspicuousa | 23 (41) | 8.9 (2.8) | 69.3 (1.4) | ||
| Conspicuousb | 33 (59) | 9.9 (3.1) | 63.8 (1.5) | ||
| Prior skin cancer | .13c | .71c | |||
| Yes | 22 (39) | 9.2 (4.2) | 63.2 (1.5) | ||
| No | 33 (61) | 9.7 (2.3) | 67.8 (1.4) | ||
| History of anxiety/depression | .91c | .91c | |||
| Yes | 12 (21) | 10.9 (2.5) | 65.1 (1.5) | ||
| No | 44 (79) | 9.2 (3.1) | 66.2 (1.4) | ||
| Relationship status | .78d | .29d | |||
| Married/partnered | 44 (79) | 9.1 (3.1) | 66.2 (1.5) | ||
| Single | 3 (5) | 23.8 (1.2) | 47.4 (1.1) | ||
| Divorced | 2 (4) | 10.1 (3.5) | 65.9 (1.6) | ||
| Widowed | 7 (13) | 8.4 (3.2) | 74.2 (1.2) | ||
| Biopsy subtype | .42d | .64d | |||
| Melanoma in situ | 39 (70) | 8.8 (3.0) | 64.5 (1.5) | ||
| Invasive melanoma | 17 (30) | 11.4 (3.1) | 69.3 (1.4) | ||
| Repair type | .78d | .99d | |||
| Secondary | 1 (2) | NA | NA | NA | NA |
| Primary | 15 (27) | 10.7 (2.2) | .78e | 66.2 (1.5) | .99e |
| Flap | 16 (29) | 9.1 (4.2) | 61.5 (1.5) | ||
| Skin graft | 9 (16) | 11.7 (3.0) | 69.0 (1.4) | ||
| Flap + graft | 7 (13) | 6.9 (2.2) | 72.2 (1.3) | ||
| Skin substitute (ie, xenograft) | 6 (11) | 8.2 (3.6) | 64.8 (1.5) | ||
| Not recorded | 2 (4) | 15.1 (4.1) | 69.9 (1.5) | ||
Abbreviation: NA, not applicable.
Inconspicuous tumors were those located in any other region of the head and neck.
Conspicuous tumors were those that lay in the region of the face bounded by the canthi laterally, the hairline superiorly, and the lower edge of the mandible inferiorly.
Unpaired, 2-tailed t tests were used to assess the differences in Skindex-16 and SCI scores between groups.
One-way analysis of variance was used to assess the differences in Skindex-16 and SCI scores between groups.
The P values associated with repair type were estimated with an analysis of variance with 6 between-groups degrees of freedom. There is only one P value for Skindex and one P value for SCI associated with the ANOVA for these comparisons.
In total, 41 questionnaires (73%) at perioperative, 49 (88%) at 6-month postoperative, and 41 (73%) at 1-year postoperative time points were completed. Differences in mean Skindex-16 and SCI scores at these time points are shown in Table 2. At baseline, the emotions subscale of both PROM instruments showed the greatest changes in HRQoL. On the Skindex-16 questionnaire, the total score was statistically significantly worse at the perioperative time point compared with baseline (26.3 vs 14.6; P < .001) but improved at the 1-year postoperative time point from the baseline (9.4 vs 14.6; P = .04). The total SCI score improved from baseline to 6-month (68.8 vs 81.5; P < .001) and 1-year (68.8 vs 82.9; P < .001) postoperative follow-up. The emotions subscale scores on the Skindex-16 questionnaire showed the greatest improvement from baseline to 1-year follow-up levels (26.6 vs 15.3; P < .001), and so did the appearance subscale scores on the SCI questionnaire (64.0 vs 84.6; P < .001). Items with the worst scores at baseline were worry about skin condition in the Skindex-16 questionnaire and worry about future cancers in the SCI questionnaire.
Table 2. Differences in Skindex-16 and Skin Cancer Index Scores Over Time.
| Questionnaire Subscale | Skindex-16 Mean Scorea | Skin Cancer Index Mean Scoreb | ||||||
|---|---|---|---|---|---|---|---|---|
| Baseline | Perioperative | 6-mo Postoperative | 1-y Postoperative | Baseline | Perioperative | 6-mo Postoperative | 1-y Postoperative | |
| Total | 14.6 | 26.3c | 10.9 | 9.4d | 68.8 | 68.2 | 81.5c | 82.9c |
| Subscale | ||||||||
| Symptoms | 7.1 | 23.6c | 8.5 | 5.0 | NA | NA | NA | NA |
| Emotions | 26.6 | 26.3 | 17.2c | 15.3c | 58.2 | 61.8 | 70.2c | 70.9c |
| Functioning | 10.1 | 28.7c | 7.2 | 7.7 | NA | NA | NA | NA |
| Social | NA | NA | NA | NA | 84.3 | 80.2 | 91.8c | 92.9c |
| Appearance | NA | NA | NA | NA | 64.0 | 62.3 | 82.3c | 84.6c |
Abbreviation: NA, not applicable.
Skindex-16 consisted of 16 items across 3 subscales (symptoms, emotions, and functioning). Responses were on a 7-box Likert-type scale anchored by the phrases never bothered to always bothered. Lower scores indicated better health-related quality of life.
Skin Cancer Index consisted of 15 items across 3 subscales (emotions, social, and appearance). Responses were on a 5-box scale ranging from very much to not at all. Higher scores indicated better health-related quality of life. For both patient-reported outcome measure instruments, all scores were transformed to a linear scale from 0 to 100. Total score reflected the mean of the 3 subscales.
P < .001.
P = .04.
Changes in Skindex-16 and SCI scores by participant sex and age over time are presented in the Figure. Female patients had worse HRQoL at baseline on both questionnaires. However, this difference diminished over time. At the 1-year postoperative time point, female and male patients had nearly the same mean SCI scores. Patients younger than 65 years had statistically significantly higher scores (worse HRQoL) on the Skindex-16 questionnaire at nearly all time points. On the SCI questionnaire, patients aged 65 years or older had statistically significantly lower scores (worse HRQoL) at all time points.
Figure. Plot of Skindex-16 and Skin Cancer Index (SCI) Scores.
Shown are the predicted Skindex-16 (A) and SCI (B) scores by sex as well as the predicted Skindex-16 (C) and SCI (D) scores by age at each evaluation time point, using the random-effects model. All models were adjusted for patient age. For the Skindex-16 questionnaire, lower scores indicated better quality of life (QoL), whereas for the SCI questionnaire, higher scores reflected better QoL. A P < .05 denoted a statistically significant difference in scores by age.
Discussion
This study demonstrated a decline in perioperative HRQoL that subsequently improved to above baseline levels by 6 months and 1 year after the surgical procedure in patients with Tis or T1a melanoma of the head and neck. The Skindex-16 and SCI questionnaires addressed different aspects of HRQoL at various time points, providing a more comprehensive picture of change in HRQoL than was previously understood. Skindex-16 responses reflected the perioperative symptomatic and social changes in HRQoL, whereas the SCI responses highlighted the long-term improvement in HRQoL, particularly regarding appearance. Understanding the unique HRQoL changes in these patients is crucial in developing effective counseling and follow-up interventions for a growing patient population.
At baseline, the emotions subscale of both PROMs showed the greatest change in HRQoL, which is consistent with previous findings demonstrating elevated psychological distress in patients with melanoma.10,11 A study of 520 patients with melanoma found that most of the identified distress was emotional in nature (eg, worry, sadness). Furthermore, elevated levels of distress were not associated with tumor stage, an interesting finding given the substantial variability in stage-dependent prognosis.12 The results from this study support previous findings of patients with melanoma experiencing considerable emotional distress largely associated with fear or worry about future occurrence or recurrence of skin cancer.10,12,13
Both PROM instruments revealed younger (<65 years) and female patients to have statistically significantly worse HRQoL at baseline. A previous study found a greater percentage of women, compared with men, with clinically elevated levels of anxiety about their melanoma diagnosis.6 In another study, younger and female patients had significantly worse SCI scores before, 1 to 2 weeks after, and 3 months after a Mohs micrographic operation.14 In our study, the difference in scores by sex diminished over time, whereas the difference in scores by age remained through the first postoperative year. Further research is needed into HRQoL implications, but they may be most pronounced for female patients in the early treatment period and may continue in the long term for younger patients. Interestingly, the visibility of the tumor and the repair type, which may affect scarring, were not associated with substantial changes in HRQoL. The overall long-term improvement in HRQoL shows that, despite the potential for scarring, removal of the tumor is a substantial factor in this improvement. Future studies using multimodule, skin cancer–specific PROM instruments, such as the FACE-Q Skin Cancer Module,15 may capture important disease- and treatment-associated concerns in this patient population.
Limitations
Limitations of this study include the relatively small sample size from a single tertiary care center. In addition, measurement bias may have been present in patients who completed the questionnaires in-office, although we took efforts to diminish this bias. Lastly, we only studied patients who underwent surgical excision; future studies may assess the changes associated with different treatment modalities.
Conclusions
Patients who underwent surgical treatment for early-stage head and neck melanoma reported an overall long-term improvement in HRQoL, which may indicate that surgical treatment of the tumor is associated with better HRQoL. A greater change in HRQoL was seen in women and those younger than 65 years. Physicians may use this information to tailor counseling for patients.
References
- 1.Gershenwald JE, Scolyer RA, Hess KR, et al. ; for members of the American Joint Committee on Cancer Melanoma Expert Panel and the International Melanoma Database and Discovery Platform . Melanoma staging: evidence-based changes in the American Joint Committee on Cancer eighth edition cancer staging manual. CA Cancer J Clin. 2017;67(6):472-492. doi: 10.3322/caac.21409 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Connolly KL, Nijhawan RI, Dusza SW, Busam KJ, Nehal KS. Time to local recurrence of lentigo maligna: implications for future studies. J Am Acad Dermatol. 2016;74(6):1247-1248. doi: 10.1016/j.jaad.2016.02.1150 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.de Vries K, Greveling K, Prens LM, et al. Recurrence rate of lentigo maligna after micrographically controlled staged surgical excision. Br J Dermatol. 2016;174(3):588-593. doi: 10.1111/bjd.14325 [DOI] [PubMed] [Google Scholar]
- 4.Moyer JS, Rudy S, Boonstra PS, et al. Efficacy of staged excision with permanent section margin control for cutaneous head and neck melanoma. JAMA Dermatol. 2017;153(3):282-288. doi: 10.1001/jamadermatol.2016.4603 [DOI] [PubMed] [Google Scholar]
- 5.Cornish D, Holterhues C, van de Poll-Franse LV, Coebergh JW, Nijsten T. A systematic review of health-related quality of life in cutaneous melanoma. Ann Oncol. 2009;20(suppl 6):vi51-vi58. doi: 10.1093/annonc/mdp255 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Al-Shakhli H, Harcourt D, Kenealy J. Psychological distress surrounding diagnosis of malignant and nonmalignant skin lesions at a pigmented lesion clinic. J Plast Reconstr Aesthet Surg. 2006;59(5):479-486. doi: 10.1016/j.bjps.2005.01.010 [DOI] [PubMed] [Google Scholar]
- 7.Balch CM, Gershenwald JE, Atkins MB, Buzaid AC, Cascinelli N, Cochrain AJ. AJCC Cancer Staging Manual. 7th ed New York, NY: Springer International Publishing; 2010:325-346. [Google Scholar]
- 8.Chren MM, Lasek RJ, Sahay AP, Sands LP. Measurement properties of Skindex-16: a brief quality-of-life measure for patients with skin diseases. J Cutan Med Surg. 2001;5(2):105-110. doi: 10.1177/120347540100500202 [DOI] [PubMed] [Google Scholar]
- 9.Rhee JS, Matthews BA, Neuburg M, Logan BR, Burzynski M, Nattinger AB. Validation of a quality-of-life instrument for patients with nonmelanoma skin cancer. Arch Facial Plast Surg. 2006;8(5):314-318. doi: 10.1001/archfaci.8.5.314 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Boyle DA. Psychological adjustment to the melanoma experience. Semin Oncol Nurs. 2003;19(1):70-77. doi: 10.1053/sonu.2003.50006 [DOI] [PubMed] [Google Scholar]
- 11.Kasparian NA, McLoone JK, Butow PN. Psychological responses and coping strategies among patients with malignant melanoma: a systematic review of the literature. Arch Dermatol. 2009;145(12):1415-1427. doi: 10.1001/archdermatol.2009.308 [DOI] [PubMed] [Google Scholar]
- 12.Loquai C, Scheurich V, Syring N, et al. Screening for distress in routine oncological care-a survey in 520 melanoma patients. PLoS One. 2013;8(7):e66800. doi: 10.1371/journal.pone.0066800 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Tesio V, Ribero S, Castelli L, et al. Psychological characteristics of early-stage melanoma patients: a cross-sectional study on 204 patients. Melanoma Res. 2017;27(3):277-280. doi: 10.1097/CMR.0000000000000348 [DOI] [PubMed] [Google Scholar]
- 14.Zhang J, Miller CJ, O’Malley V, Etzkorn JR, Shin TM, Sobanko JF. Patient quality of life fluctuates before and after Mohs micrographic surgery: a longitudinal assessment of the patient experience. J Am Acad Dermatol. 2018;78(6):1060-1067. doi: 10.1016/j.jaad.2018.02.065 [DOI] [PubMed] [Google Scholar]
- 15.Lee EH, Klassen AF, Cano SJ, Nehal KS, Pusic AL. FACE-Q Skin Cancer Module for measuring patient-reported outcomes following facial skin cancer surgery. Br J Dermatol. 2018;179(1):88-94. doi: 10.1111/bjd.16671 [DOI] [PMC free article] [PubMed] [Google Scholar]
