Key Points
Question
What is the risk of death after a diagnosis of melanoma in situ (MIS) in the US?
Findings
In this population-based cohort study of 137 872 patients with MIS, the 15-year melanoma-specific survival was 98.4%. The 15-year relative survival, meaning overall survival relative to similar individuals in the population, was 112.4%.
Meaning
The MIS survival statistics in this cohort study suggest that patients with a diagnosis of MIS have an increased, but low risk of dying of melanoma and live longer overall than people in the general population; these statistics suggest significant detection of low-risk disease among health-seeking individuals.
Abstract
Importance
The incidence of melanoma in situ (MIS) is increasing more rapidly than any invasive or in situ cancer in the US. Although more than half of melanomas diagnosed are MIS, information about long-term prognosis following a diagnosis of MIS remains unknown.
Objective
To evaluate mortality and factors associated with mortality after a diagnosis of MIS.
Design, Setting, and Participants
This population-based cohort study of adults with a diagnosis of first primary MIS from 2000 to 2018 included data from the US Surveillance, Epidemiology, and End Results Program, which were analyzed from July to September 2022.
Main Outcomes and Measures
Mortality after a diagnosis of MIS was evaluated using 15-year melanoma-specific survival, 15-year relative survival (ie, compared with similar individuals without MIS), and standardized mortality ratios (SMRs). Cox regression was used to estimate hazard ratios (HRs) for death by demographic and clinical characteristics.
Results
Among 137 872 patients with a first-and-only MIS, the mean (SD) age at diagnosis was 61.9 (16.5) years (64 027 women [46.4%]; 239 [0.2%] American Indian or Alaska Native, 606 [0.4%] Asian, 344 [0.2%] Black, 3348 [2.4%] Hispanic, and 133 335 [96.7%] White individuals). Mean (range) follow-up was 6.6 (0-18.9) years. The 15-year melanoma-specific survival was 98.4% (95% CI, 98.3%-98.5%), whereas the 15-year relative survival was 112.4% (95% CI, 112.0%-112.8%). The melanoma-specific SMR was 1.89 (95% CI, 1.77-2.02); however, the all-cause SMR was 0.68 (95% CI, 0.67-0.7). Risk of melanoma-specific mortality was higher for older patients (7.4% for those 80 years or older vs 1.4% for those aged 60-69 years; adjusted HR, 8.2; 95% CI, 6.7-10.0) and patients with acral lentiginous histology results (3.3% for acral lentiginous vs 0.9% for superficial spreading; HR, 5.3; 95% CI, 2.3-12.3). Of patients with primary MIS, 6751 (4.3%) experienced a second primary invasive melanoma and 11 628 (7.4%) experienced a second primary MIS. Compared with patients without a subsequent melanoma, the risk of melanoma-specific mortality was increased for those with a second primary invasive melanoma (adjusted HR, 4.1; 95% CI, 3.6-4.6) and was decreased for those with a second primary MIS (adjusted HR, 0.7; 95% CI, 0.6-0.9).
Conclusions and relevance
The results of this cohort study suggest that patients with a diagnosis of MIS have an increased but low risk of melanoma-specific mortality and live longer than people in the general population, suggesting that there is significant detection of low-risk disease among health-seeking individuals. Factors associated with death following MIS include older age (≥80 years) and subsequent primary invasive melanoma.
This cohort study examines mortality and factors associated with mortality after a diagnosis of melanoma in situ.
Introduction
The incidence of cutaneous melanoma has rapidly increased during the past 5 decades.1 Once a rare tumor, melanoma is now the fourth most commonly diagnosed cancer in the US.2 Melanoma in situ (MIS), which is also known as stage 0 melanoma, has disproportionately contributed to the rising incidence. In 1975, the incidence rate of MIS was 0.5 per 100 000 population; it is now more than 25 per 100 000 population (a 50-fold increase).1 Melanoma in situ currently accounts for approximately half of detected melanomas.3,4 Although MIS is not generally considered life-threatening, patients with MIS are at increased risk of receiving diagnoses of subsequent melanomas (invasive and in situ); thus these individuals warrant more frequent surveillance.5,6 While MIS is regarded as a precursor lesion to invasive melanoma, the natural history of MIS and its likelihood of progression is not clear.7,8 Despite the increasing frequency of MIS diagnosis, to our knowledge, little information exists about its association with melanoma-specific or all-cause mortality.
Cancer survival statistics convey information to patients and help clinicians put diagnoses into context.9 These statistical measures can be presented in many ways. The most common and familiar measure is disease-specific survival, which is the proportion of patients who remain alive over a certain period after a diagnosis of cancer. Relative survival and standardized mortality ratios (SMRs) are less commonly used measures that provide useful information about cancer mortality compared with similar individuals in the general population without cancer. These measures can help clinicians orient patients’ expectations about prognosis, as some cancers may be associated with living longer.10 For example, patients with a diagnosis of ductal carcinoma in situ (stage 0 breast cancer) are at a slightly increased risk of death of breast cancer compared with the general population, but are at decreased risk of death overall.11,12,13 This phenomenon has been explained by the detection of indolent cancers, either incidentally or through screening, in otherwise healthy people.14
To date, mortality after a diagnosis of MIS remains unclear. It is also unknown what factors are associated with the risk of mortality in patients with MIS and how the risk compares with that of individuals in the general population. Understanding the risk factors for mortality among patients with MIS may provide information to identify specific subgroups that may be at higher risk of death and other groups at low risk that may be harmed by overdiagnosis and overtreatment.15 In this study, we aimed to evaluate mortality (and factors associated with mortality) after a diagnosis of MIS using several complementary cancer statistics: melanoma-specific survival, relative survival, and SMRs.
Methods
Data Source
This retrospective population-based cohort study was conducted using the Surveillance, Epidemiology, and End Results (SEER) Program (version 8.4.0.1; National Cancer Institute).16 Data on incident cases of melanoma were obtained from 18 SEER registries. Mortality data were obtained through the National Vital Statistics System, which provides cause-of-death information for deaths in the US.17 Because the data were deidentified and publicly available, it was deemed exempt by the institutional review board at The University of Texas at Austin, and informed consent was waived.
Case Selection
The sample included adults aged 20 years or older with a diagnosis of a first primary cutaneous MIS between 2000 and 2018. We used histologic behavior codes from The International Classification of Diseases for Oncology, 3rd edition,18 to differentiate between in situ (/2) and invasive (/3) melanoma, as described previously (eMethods in Supplement 1).1,19,20 We excluded 0.2% of MIS cases that had a recorded nonzero Breslow depth due to concerns for misclassification. eFigure 1 in the Supplement 1 provides the cohort selection flow diagram.
Further information extracted from SEER included age at diagnosis, sex, year of diagnosis (stratified by pre/post 2011, the year of introduction of novel therapy), race and ethnicity, median area-level annual household income, metropolitan status (rural or urban), marital status, tumor site, and histologic subtype. Information about household income and metropolitan status in SEER represents county-level estimates from the 2015 to 2018 American Community Survey.21 Patients who experienced a second primary cutaneous melanoma were identified using the SEER tumor sequence number, which indicates the order in which a primary tumor is discovered compared with the total number of primaries for each patient. The corresponding date of diagnosis and histologic behavior (in situ or invasive) for each second primary was recorded.
Vital status was evaluated at the time of last contact using the following SEER cause-specific death classifications: (1) alive, (2) dead due to melanoma, or (3) dead due to other causes. Survival was computed as the time (in months) from diagnosis to last follow-up or study cutoff. The study cutoff date was December 31, 2018. SEER*Stat software (version 8.4.0.1) was used to generate the case listing.
Analysis
We used several complementary measures to evaluate the primary outcome, mortality, in 2 separate cohorts of patients with MIS (eFigure 1 in Supplement 1). The first cohort included patients who had a first and only diagnosis of MIS, without any other subsequent primary melanomas. This allowed us to investigate the relationship between the diagnosis of a MIS and mortality. The second cohort included all patients with a first primary MIS, including those who later developed a second primary melanoma. This enabled us to evaluate the risk of developing a second primary melanoma (in situ or invasive) in patients with MIS and its overall association with mortality. Data were analyzed from July to September 2022.
Melanoma-Specific and Relative Survival
We compared melanoma-specific survival and relative survival at 15 years to maximize possible follow up tim,e as melanoma is known to have late recurrences 10 or more years after initial treatment.22 Melanoma-specific survival is the proportion of patients who do not die of melanoma. Melanoma-specific survival was calculated using the Kaplan-Meier method to censor for nonmelanoma causes of death. Relative survival is the overall survival of patients with MIS compared with the overall survival in similar individuals without MIS (ie, relative survival measures whether a diagnosis of MIS shortens life).10,23 Relative survival was calculated as the ratio of observed to expected overall survival in age-matched, sex-matched, and race-matched individuals in the general population; the Ederer II method was used to calculate expected survival based on US annual life tables through a survival session in the SEER*Stat software.24,25
Standardized Mortality Ratio
We calculated SMRs to compare the risk of melanoma-specific mortality following a diagnosis of MIS for individuals within the cohort compared with the general US population. To provide further context, we also calculated SMRs for the risk of death of all causes between the cohort of patients with MIS and the general population. The SMR was calculated as the ratio of observed to expected deaths as stratified by age at diagnosis; expected deaths represented the incidence-based mortality of each age group in the general population. Because patients with melanoma are generally more likely to be of non-Hispanic White race than patients in the general population, we performed a sensitivity analysis to evaluate whether the SMRs were stable when the population was restricted to non-Hispanic White individuals. The SMRs and exact 95% confidence limits for SMRs, which were derived from a Poisson regression model, were calculated using SEER*Stat software.
Hazard Ratio (HR) of Death
We performed Cox proportional hazards regression to identify factors associated with melanoma-specific mortality following a diagnosis of MIS, including the effect of a second primary melanoma (in situ or invasive). We used 2 multivariable models. The first model was restricted to patients with a first and only MIS (cohort 1) and included covariates based on a priori knowledge of associations with sociodemographic and clinical factors. The second model included patients with MIS who also experienced 1 or more subsequent primary melanomas (cohort 2). In this model, type of second primary melanoma (invasive, in situ, or no second primary) was added as the exposure of interest, and HRs for melanoma-specific mortality were adjusted for each of the covariates selected for the first model. Statistical modeling was performed using MATLAB 2021b (MathWorks). Statistical testing was 2-sided (α = .05).
Results
Analysis of First and Only Melanomas and Mortality
A total of 137 872 patients (cohort 1) with first and only primary MIS were identified. More than half were male (73 845 [53.6%]) and older than 60 years (78 036 [56.6%]), and nearly all were White (133 335 [96.7%]) (Table 1). During the 910 308 person-years of follow-up (mean [SD], 6.6 [5.1] years), 893 (0.6%) died of melanoma, and 17 327 (12.6%) died of any cause. In the multivariable analysis, older age (7.4% for ≥80 years vs 1.4% for 60-69 years; adjusted HR, 8.2; 95% CI, 6.7-10.0; P < .001), male sex (2.2% vs 1.0%; adjusted HR, 1.9; 95% CI, 1.6-2.2; P < .001), American Indian or Alaska native race (5.5% vs 1.6% for non-Hispanic White; adjusted HR, 4.3; 95% CI, 1.6-11.5; P = .004), and acral lentiginous histology (3.3% vs 0.9% superficial spreading histology; adjusted HR, 5.3; 95% CI, 2.3-12.3; P < .001) were associated with a higher risk of 15-year melanoma-specific mortality following a diagnosis of MIS (Table 1).
Table 1. Mortality and HRs for Melanoma-Specific Mortality Following a First and Only MIS by Patient Characteristics.
| Characteristic | Patients, No. (%) | 15-y Mortality, % (95% CI) | Unadjusted HR (95% CI) | P value | Adjusted HR (95% CI)a | P value |
|---|---|---|---|---|---|---|
| All patients | 137 872 | 1.6 (1.5-1.7) | NA | NA | NA | NA |
| Age at diagnosis, y | ||||||
| 20-29 | 3431 (2.5) | 0.1 (0-0.3) | 0.1 (0-0.4) | <.001 | 0.1 (0-0.4) | .001 |
| 30-39 | 9095 (6.6) | 0.1 (0-0.2) | 0.1 (0-0.2) | <.001 | 0.1 (0.1-0.3) | <.001 |
| 40-49 | 17 544 (12.7) | 0.3 (0.1-0.4) | 0.2 (0.1-0.3) | <.001 | 0.2 (0.1-0.3) | <.001 |
| 50-59 | 29 662 (21.5) | 0.5 (0.4-0.7) | 0.4 (0.3-0.6) | <.001 | 0.5 (0.3-0.6) | <.001 |
| 60-69 | 35 088 (25.4) | 1.4 (1.1-1.7) | 1 [Reference] | NA | 1 [Reference] | NA |
| 70-79 | 26 619 (19.3) | 4.4 (3.7-5) | 3.3 (2.7-4) | <.001 | 3.3 (2.7-4.1) | <.001 |
| ≥80 | 16 433 (11.9) | 7.4 (5.9-8.9) | 8.3 (6.8-10.1) | <.001 | 8.2 (6.7-10) | <.001 |
| Sex | ||||||
| Female | 64 027 (46.4) | 1 (0.9-1.1) | 1 [Reference] | NA | 1 [Reference] | NA |
| Male | 73 845 (53.6) | 2.2 (2-2.4) | 2.2 (1.9-2.5) | <.001 | 1.9 (1.6-2.2) | <.001 |
| Year of diagnosis | ||||||
| 2000-2010 | 59 756 (43.3) | 1.6 (1.5-1.8) | 1 [Reference] | NA | 1 [Reference] | NA |
| 2011-2018 | 78 116 (56.7) | 0.6 (0.5-0.8) | 0.8 (0.7-1) | .03 | 0.7 (0.6-0.9) | <.001 |
| Race and ethnicity | ||||||
| American Indian or Alaska Native | 239 (0.2) | 5.5 (0-11.1) | 2.7 (1-7.3) | .05 | 4.3 (1.6-11.5) | .004 |
| Asian | 606 (0.4) | 1.8 (0.2-3.4) | 1.2 (0.5-2.9) | .7 | 1.4 (0.6-3.3) | .47 |
| Black | 344 (0.2) | 0.4 (0-1.2) | 0.4 (0.1-3.1) | .4 | 0.3 (0-2.4) | .27 |
| Hispanic | 3348 (2.4) | 1.2 (0.6-1.8) | 0.9 (0.5-1.4) | .51 | 1 (0.6-1.6) | .95 |
| White | 133 335 (96.7) | 1.6 (1.5-1.7) | 1 [Reference] | NA | 1 [Reference] | NA |
| Annual household income, $ | ||||||
| <50 000 | 15 020 (10.9) | 2 (1.3-2.7) | 1 [Reference] | NA | 1 [Reference] | NA |
| 50 000-59 999 | 18 946 (13.7) | 1.4 (1.1-1.7) | 1 (0.8-1.3) | .98 | 1 (0.8-1.4) | .88 |
| 60 000-69 999 | 36 177 (26.2) | 1.7 (1.5-2) | 1.1 (0.9-1.5) | .34 | 1.2 (0.9-1.6) | .18 |
| >70 000 | 67 687 (49.1) | 1.5 (1.3-1.7) | 1 (0.8-1.3) | .99 | 1.1 (0.8-1.5) | .41 |
| Unknown | 42 (0.03) | 4.2 (0-12.2) | 3.5 (0.5-25.4) | .21 | 3.3 (0.4-24) | .24 |
| Residence | ||||||
| Rural | 14 394 (10.4) | 1.7 (1.2-2.2) | 1 [Reference] | NA | 1 [Reference] | NA |
| Urban | 123 478 (89.6) | 1.6 (1.4-1.7) | 1 (0.8-1.3) | .97 | 1 (0.8-1.3) | .84 |
| Relationship status | ||||||
| Married | 56 294 (40.8) | 1.4 (1.2-1.6) | 1 [Reference] | NA | 1 [Reference] | NA |
| Never married | 10 228 (7.4) | 1.4 (1-1.8) | 1 (0.7-1.3) | .82 | 1.6 (1.3-2.2) | <.001 |
| Separated/widowed | 10 540 (7.6) | 3.2 (2.5-3.9) | 2.4 (1.9-2.9) | <.001 | 1.5 (1.2-1.9) | <.001 |
| Unknown | 60 810 (44.1) | 1.6 (1.4-1.8) | 1.1 (0.9-1.3) | .26 | 1.1 (1-1.3) | .08 |
| Primary site | ||||||
| Extremities | 53 712 (39) | 0.9 (0.8-1.1) | 1 [Reference] | NA | 1 [Reference] | NA |
| Face or ears | 35 462 (25.7) | 2.5 (2.2-2.9) | 2.7 (2.3-3.1) | <.001 | 1.4 (1.2-1.7) | <.001 |
| Scalp or neck | 11 255 (8.2) | 3 (2.2-3.7) | 2.6 (2.1-3.3) | <.001 | 1.6 (1.2-2) | <.001 |
| Trunk | 36 840 (26.7) | 1.4 (1.2-1.6) | 1.3 (1.1-1.6) | .01 | 1.2 (1-1.5) | .08 |
| Unspecified | 603 (0.4) | 1.6 (0.1-3.1) | 2 (0.8-4.9) | .12 | 1.6 (0.7-3.9) | .3 |
| Histologic subtype | ||||||
| Superficial spreading | 9565 (6.9) | 0.9 (0.5-1.2) | 1 [Reference] | NA | 1 [Reference] | NA |
| Acral lentiginous | 372 (0.3) | 3.3 (0.8-5.8) | 5.3 (2.3-11.9) | <.001 | 5.3 (2.3-12.3) | <.001 |
| Lentigo maligna | 33 616 (24.4) | 2.3 (2-2.6) | 3 (2.1-4.2) | <.001 | 1.1 (0.8-1.6) | .61 |
| Not otherwise specified | 94 216 (68.3) | 1.4 (1.3-1.6) | 1.8 (1.3-2.6) | <.001 | 1.3 (0.9-1.8) | .14 |
| Other | 103 (0.1) | 2.8 (0-6.8) | 5.4 (1.3-22.3) | .02 | 4.1 (1-17) | .05 |
Abbreviations: MIS, melanoma in situ; NA, not applicable; HR, hazard ratio
Hazard ratios were adjusted for age at diagnosis, sex, year of diagnosis, race and ethnicity, income, residence type, relationship status, primary site, and histologic subtype.
At 15 years, melanoma-specific survival was 98.4% (95% CI, 98.3%-98.5%), whereas relative survival was 112.4% (95% CI, 112.0%-112.8%) (Figure 1), meaning that patients with a diagnosis of MIS lived longer than the general population without a history of MIS. Yearly, interval survival data are shown in eTable 1 in Supplement 1.
Figure 1. Fifteen-Year Relative and Melanoma-Specific Survival of Patients With a First and Only Melanoma in Situ (MIS).
Relative survival compares the overall survival of people with a diagnosis of MIS with the overall survival of age-matched, sex-matched, and race and ethnicity–matched individuals without a diagnosis of MIS.
In the SMR analysis, melanoma-specific mortality after a diagnosis of a first and only MIS was 1.89 (95% CI, 1.77-2.02) times greater than that of the general population, although the 15-year mortality risk was 1.6% (95% CI, 1.5-1.7) (Table 2). The risk of mortality of all causes for patients with MIS was significantly lower than that of the general population (SMR, 0.68; 95% CI, 0.67-0.7). There was minimal variation in SMRs as stratified by age at diagnosis. Further, the SMRs were similar when the population was restricted to White individuals, suggesting that the study results were not sensitive to White race (eTable 2 in Supplement 1).
Table 2. SMRs for Melanoma and All Causes Following a Diagnosis of a First and Only Melanoma in Situ by Age at Diagnosis.
| Age at diagnosis, y | Person-y | Melanoma-specific | All-cause | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Deaths, No. | Cumulative mortality, % | 15-y Mortality, % (95% CI) | SMR (95% CI)a | Deaths, No. | Cumulative mortality, % | 15-y Mortality, % (95% CI) | SMR (95% CI)a | ||||
| Observed | Expected | Observed | Expected | ||||||||
| 20-29 | 30 415.5 | 2 | 0.4 | 0.1 | 0.1 (0-0.3) | 5.56 (0.52-15.92) | 19 | 32 | 0.6 | 1.2 (0.6-1.8) | 0.59 (0.36-0.93) |
| 30-39 | 77 660.7 | 6 | 1.6 | 0.1 | 0.1 (0-0.2) | 3.74 (1.35-7.33) | 66 | 157.8 | 0.7 | 1.3 (0.9-1.6) | 0.42 (0.32-0.53) |
| 40-49 | 148 444.3 | 20 | 11.1 | 0.1 | 0.3 (0.1-0.4) | 1.8 (1.1-2.67) | 240 | 690 | 1.4 | 2.7 (2.4-3.1) | 0.35 (0.31-0.39) |
| 50-59 | 217 231.7 | 67 | 36.2 | 0.2 | 0.5 (0.4-0.7) | 1.85 (1.43-2.32) | 803 | 2153 | 2.7 | 6.6 (6-7.1) | 0.37 (0.35-0.4) |
| 60-69 | 213 575.1 | 141 | 99.7 | 0.4 | 1.4 (1.1-1.7) | 1.41 (1.19-1.66) | 2298 | 4616.7 | 6.6 | 20.2 (19.3-21.2) | 0.5 (0.48-0.52) |
| 70-79 | 150 421.4 | 319 | 158.4 | 1.2 | 4.4 (3.7-5) | 2.01 (1.8-2.24) | 5805 | 8498.2 | 21.8 | 58.2 (56.9-59.5) | 0.68 (0.67-0.7) |
| ≥80 | 72 559.5 | 338 | 155.3 | 2.1 | 7.4 (5.9-8.9) | 2.18 (1.95-2.41) | 8096 | 9152 | 49.3 | 92.6 (91.6-93.5) | 0.88 (0.87-0.9) |
| All | 910 308.1 | 893 | 471.6 | 0.7 | 1.6 (1.5-1.7) | 1.89 (1.77-2.02) | 17327 | 25299.6 | 12.6 | 27.1 (26.7-27.6) | 0.68 (0.67-0.7) |
Abbreviations: SMR, standardized mortality ratio.
Standardized mortality ratios were computed for the entire cohort of patients with a first and only melanoma in situ (n = 137 872). Cumulative mortality was computed as the number of observed deaths divided by the number of at-risk individuals at the start of the study period.
Analysis of Second Primary Melanomas and Mortality
Among a total of 156 251 patients (cohort 2) with first primary MIS, 18 379 patients (11.8%) also experienced a second primary melanoma, of which 6751 (36.7%) were invasive and 11 628 (63.3%) were in situ. eTable 3 in Supplement 1 provides demographic characteristics for the study sample. At 15 years, the cumulative incidence of second primary invasive melanoma was 8.9% (95% CI, 8.6%-9.2%), while that of second primary MIS was 14.1% (95% CI, 13.8%-14.5%) (eFigure 2 in Supplement 1).
At 15 years, for patients who experienced a second primary MIS, melanoma-specific survival was 98.2% (95% CI, 97.6%-98.5%), while relative survival was 126.7% (95% CI, 125.5%-128.0%). For patients who experienced a second primary invasive melanoma, melanoma-specific survival was 91.1% (95% CI, 90.0%-92.1%), while relative survival was 116.7% (95% CI, 115%-118.4%) (Figure 2).
Figure 2. Fifteen-Year Relative and Melanoma-Specific Survival of Patients With Melanoma in Situ (MIS) Who Experienced a Second Primary Melanoma by Type of Second Primary.
Relative survival compares the overall survival of people with a diagnosis of MIS with the overall survival of age-matched, sex-matched, and race and ethnicity–matched individuals without a diagnosis of MIS.
In the SMR analysis, the melanoma-specific mortality after a diagnosis of MIS was 4.27 times greater than that of the general population (SMR, 4.27; 95% CI, 4.07-4.48) (eTable 4 in Supplement 1). However, the risk of death of all causes for the same cohort remained significantly lower than that of the general population (SMR, 0.73; 95% CI, 0.72-0.74).
Compared with patients who had a first and only MIS, the risk of melanoma-specific mortality was increased for those who experienced a second primary invasive melanoma (adjusted HR, 4.1; 95% CI, 3.6-4.6; P < .001) and was slightly decreased for patients who experienced a second primary MIS (adjusted HR, 0.7; 95% CI, 0.6-0.9; P < .001) (Table 3). The Kaplan-Meier curves for melanoma-specific survival by type of second primary are shown in eFigure 3 in Supplement 1.
Table 3. Mortality and HRs for Melanoma-Specific Mortality Following First Primary MIS by Type of Second Primary.
| Second primary | Patients, No. (%) | 15-y Mortality, % (95% CI) | Univariate HR (95% CI) | P value | Multivariate HR (95% CI)a | P value |
|---|---|---|---|---|---|---|
| No subsequent primary | 137 872 (88.2) | 1.6 (1.5-1.7) | 1 [Reference] | NA | 1 [Reference] | NA |
| Invasive melanoma | 6751 (4.3) | 8.9 (7.9-9.9) | 5.6 (5-6.4) | <.001 | 4.1 (3.6-4.6) | <.001 |
| MIS | 11 628 (7.4) | 1.8 (1.4-2.3) | 0.9 (0.7-1.1) | .39 | 0.7 (0.6-0.9) | <.001 |
Abbreviations: HR; hazard ratio, MIS, melanoma in situ; NA, not applicable.
Hazard ratios were adjusted for age at diagnosis, sex, year of diagnosis, race and ethnicity, income, residence type, relationship status, primary site, and histologic subtype.
Discussion
Given that half of patients with melanoma receive a diagnosis of MIS, it is important that accurate information about prognosis is available to patients and clinicians. To understand the prognosis of patients with a diagnosis of MIS, we investigated 3 cancer survival statistics: disease-specific survival, relative survival, and SMRs. We found that at 15 years, patients with a diagnosis of MIS had a 98.4% disease-specific survival rate, which was slightly higher than patients with a diagnosis of thin (≤1.0 mm) invasive melanoma in an Australian cohort (96.7%) or in patients with stage I melanoma in SEER (95.1%).16,26 However, the relative survival was greater than 100%, meaning that these patients were living longer compared with age-matched, sex-matched, and race and ethnicity–matched individuals in the US population for up to 15 years after diagnosis. Similarly, SMRs suggested that while patients with MIS had slightly increased mortality of melanoma than expected (based on standard life tables for the US population), their mortality of all causes was lower than expected. Clinical features associated with increased risk of 15-year melanoma-specific mortality included male sex, older age, acral lentiginous histology, and development of a separate (primary) invasive melanoma (which has been previously described in the literature).27,28,29 Development of a second primary MIS was associated with a further decrease in the risk of melanoma-specific mortality and further increases in relative survival.
Taken together, these cancer survival statistics that demonstrated increased overall survival of patients with MIS compared with the general population suggest that in most cases, diagnosis of MIS is a risk factor for future melanoma diagnosis and rarely portends adverse outcomes. However, we are unable to predict which cases of MIS may progress to invasive melanoma, so all are treated. This study’s findings potentially lend further credence to the idea that MIS is not an obligate precursor lesion to invasive melanoma.1,4,8 Furthermore, interpreting MIS as cancer may contribute to overdiagnosis, the detection of very low-risk disease that would not cause symptoms or death if left alone. Cancer overdiagnosis is often associated with screening or incidental detection of cancer among healthy individuals with access to health care. In this context, relative survival statistics may be reassuring to patients, showing that the health-related decisions they made before receiving their diagnosis may have a greater positive association with their overall survival than the diagnosis itself. Although this finding may help reduce feelings of anxiety and depression that are associated with an initial diagnosis, it is also important to consider the other potential harmful effects of a MIS diagnosis, namely the psychological effects of being labeled with cancer, including fear of cancer recurrence and the specter of lifelong skin examinations.30
The fear of cancer recurrence is an important consideration, given that prior studies have established that patients with MIS are at heightened risk of developing subsequent cutaneous melanomas (in situ and invasive) at different sites on the body.31 However, these studies have not established how this risk translates to mortality. In the present study, we demonstrated that patients who even develop a second primary melanoma (in situ or invasive) also had a relative survival greater than 100% at 15 years after receiving their initial diagnosis of MIS. While the risk of experiencing a second melanoma is concerning for patients with MIS, it is important to understand that this risk may not necessarily translate to a shorter life. We also examined disease-specific survival for patients who ultimately developed a second primary. Patients who developed a second MIS had a decreased risk of melanoma-specific mortality compared with those with a single MIS, further suggesting that an MIS diagnosis (and even multiple MIS diagnoses) may be a marker of good overall health or health-seeking behavior. Nevertheless, patients who developed a second primary invasive melanoma after their initial MIS, despite surviving longer overall compared with the general population, were more likely to die of melanoma than those with a single MIS. Distinguishing the minority of individuals who are at relatively higher risk of dying of melanoma may identify patients who will benefit from more aggressive follow-up while reducing harm associated with unnecessary treatment of patients whose survival is unlikely to be affected by their MIS.
In the present study, we identified the following risk factors for melanoma-specific mortality after a diagnosis of MIS: race, older age, male sex, and acral histology. These risk factors are similar for patients with invasive melanoma. Notably, American Indian and Alaska Native race and ethnicity was the only racial and ethnic category that was associated with higher 15-year melanoma-specific mortality (compared with White individuals) following a diagnosis of MIS. Understanding these risk factors may help stratify patients into high-risk groups that could be followed more frequently after a diagnosis of MIS vs low-risk groups that may not need to be followed at all.15 Currently, practice and guidelines regarding follow up after MIS vary globally, with some guidelines from the US, such as the National Comprehensive Cancer Network and American Academy of Dermatology, recommending annual screening for a lifetime and other international organizations not advocating for follow up screening, such as the British Association of Dermatologists.32,33,34,35 Others in Europe and Australia make no recommendations.36,37
Some may posit that the risk of death from MIS is theoretically 0 given that the cancerous cells are restricted to the epidermis. One possible explanation for the observed mortality of MIS is that some cases may represent misdiagnosed invasive melanomas due to sampling errors or histologic ambiguity. Alternatively, some previously invasive melanomas that evaded detection may undergo spontaneous regression, leaving behind only MIS at the time of diagnosis. Nonetheless, the present study data suggest that mortality of MIS is extremely rare.
Limitations
Although the size and longitudinal design of the SEER registry enabled us to provide mortality estimates, the registry has several inherent limitations. A few cases in the SEER registry may have inconsistent coding of staging data.38 Further, it is possible that a more thorough pathology review could have led to the reclassification of some cases of MIS as invasive melanoma or vice versa.39 To reduce potential misclassification, we excluded a minority of cases that had a nonzero Breslow depth. Although we included household income and metropolitan status in our models, these county-level measures alone are not ideal indicators of socioeconomic status or health care access. Due to the limited availability of other social or clinical covariates (eg, insurance, health care utilization, overall health status), conclusions about health care–seeking behavior of patients from evaluations of relative survival and SMRs may only be inferred. Finally, accurate calculation of disease-specific survival rates requires correct classification of the cause of death, and accurate life tables for the background population are necessary for relative survival calculations.
Conclusions
The melanoma cancer survival statistics reported in this cohort study are potentially important for patients and clinicians to accurately understand and put into perspective the risks associated with receiving a diagnosis of MIS. Patients with a diagnosis of MIS have an increased but low risk of melanoma-specific mortality and live longer than people in the general population, suggesting significant detection of low-risk disease among health-seeking individuals.
eMethods. Identification of cases of melanoma in situ
eFigure 1. Cohort flow diagram
eTable 1. Relative and disease-specific survival following melanoma in situ
eTable 2. Sensitivity analysis: Standardized mortality ratios for melanoma and all-causes following a diagnosis of a first and only melanoma in situ, by age at diagnosis, restricted to non-Hispanic White persons
eTable 3. Characteristics of patients who experienced a first primary MIS, by type of second primary
eFigure 2. Cumulative incidence of second primary melanoma among patients with a prior first primary melanoma in situ, by type of second primary
eTable 4. Standardized mortality ratios for disease-specific and all-cause mortality following a diagnosis of melanoma in situ and subsequent primary melanomas, by age at diagnosis
eFigure 3. Melanoma-specific survival following MIS by type of second primary melanoma
Data sharing statement
References
- 1.Welch HG, Mazer BL, Adamson AS. The rapid rise in cutaneous melanoma diagnoses. N Engl J Med. 2021;384(1):72-79. doi: 10.1056/NEJMsb2019760 [DOI] [PubMed] [Google Scholar]
- 2.National Cancer Institute . 1975-2018: SEER. Accessed November 29, 2022. https://seer.cancer.gov/archive/csr/1975_2018/browse_csr.php?sectionSEL=1&pageSEL=sect_01_table.01
- 3.Higgins HW II, Lee KC, Galan A, Leffell DJ. Melanoma in situ: part I: epidemiology, screening, and clinical features. J Am Acad Dermatol. 2015;73(2):181-190. doi: 10.1016/j.jaad.2015.04.014 [DOI] [PubMed] [Google Scholar]
- 4.Olsen CM, Pandeya N, Rosenberg PS, Whiteman DC. Incidence of in situ vs invasive melanoma: testing the “obligate precursor” Hypothesis. J Natl Cancer Inst. 2022;114(10):1364-1370. doi: 10.1093/jnci/djac138 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Balamurugan A, Rees JR, Kosary C, Rim SH, Li J, Stewart SL. Subsequent primary cancers among men and women with in situ and invasive melanoma of the skin. J Am Acad Dermatol. 2011;65(5)(suppl 1):S69-S77. doi: 10.1016/j.jaad.2011.04.033 [DOI] [PubMed] [Google Scholar]
- 6.Wassberg C, Thörn M, Yuen J, Hakulinen T, Ringborg U. Cancer risk in patients with earlier diagnosis of cutaneous melanoma in situ. Int J Cancer. 1999;83(3):314-317. doi: [DOI] [PubMed] [Google Scholar]
- 7.Shain AH, Yeh I, Kovalyshyn I, et al. The genetic evolution of melanoma from precursor lesions. N Engl J Med. 2015;373(20):1926-1936. doi: 10.1056/NEJMoa1502583 [DOI] [PubMed] [Google Scholar]
- 8.Elder DE. Obligate and potential precursors of melanoma. J Natl Cancer Inst. 2022;114(10):1320-1322. doi: 10.1093/jnci/djac139 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Mariotto AB, Noone AM, Howlader N, et al. Cancer survival: an overview of measures, uses, and interpretation. J Natl Cancer Inst Monogr. 2014;2014(49):145-186. doi: 10.1093/jncimonographs/lgu024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Henson DE, Ries LA. The relative survival rate. Cancer. 1995;76(10):1687-1688. doi: [DOI] [PubMed] [Google Scholar]
- 11.Narod SA, Iqbal J, Giannakeas V, Sopik V, Sun P. Breast cancer mortality after a diagnosis of ductal carcinoma in situ. JAMA Oncol. 2015;1(7):888-896. doi: 10.1001/jamaoncol.2015.2510 [DOI] [PubMed] [Google Scholar]
- 12.Giannakeas V, Sopik V, Narod SA. Association of a diagnosis of ductal carcinoma in situ with death from breast cancer. JAMA Netw Open. 2020;3(9):e2017124. doi: 10.1001/jamanetworkopen.2020.17124 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Marcadis AR, Marti JL, Ehdaie B, Hakimi AA, Davies L, Morris LGT. Characterizing relative and disease-specific survival in early-stage cancers. JAMA Intern Med. 2020;180(3):461-463. doi: 10.1001/jamainternmed.2019.6120 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Welch HG, Kramer BS, Black WC. Epidemiologic signatures in cancer. N Engl J Med. 2019;381(14):1378-1386. doi: 10.1056/NEJMsr1905447 [DOI] [PubMed] [Google Scholar]
- 15.Eguchi MM, Elder DE, Barnhill RL, et al. Prognostic modeling of cutaneous melanoma stage I patients using cancer registry data identifies subsets with very-low melanoma mortality. Cancer. 2023;129(1):89-97. Published online 2022. doi: 10.1002/cncr.34490 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.National Cancer Institute Surveillance Research Program . Surveillance, Epidemiology, and End Results Program. Accessed September 28, 2022. https://seer.cancer.gov/
- 17.US Centers for Disease Control and Prevention . National vital statistics system. Accessed September 28, 2022. https://www.cdc.gov/nchs/nvss/index.htm
- 18.Jack A, Percy CL, Sobin L, Whelan S. International Classification of Diseases for Oncology: ICD-O. World Health Organization; 2000. [Google Scholar]
- 19.Mocellin S, Nitti D. Cutaneous melanoma in situ: translational evidence from a large population-based study. Oncologist. 2011;16(6):896-903. doi: 10.1634/theoncologist.2010-0340 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Daniel VT, Crawford A, Kiefe CI, Mahmoud BH. Recurrence and mortality of melanoma in situ of the trunk or extremities: a surveillance, epidemiology, and end results analysis. Dermatol Surg. 2021;47(1):1-5. doi: 10.1097/DSS.0000000000002417 [DOI] [PubMed] [Google Scholar]
- 21.US Census Bureau . American community survey 5-year data (2009-2018). Accessed July 14, 2022. https://www.census.gov/data/developers/data-sets/acs-5year.html
- 22.Faries MB, Steen S, Ye X, Sim M, Morton DL. Late recurrence in melanoma: clinical implications of lost dormancy. J Am Coll Surg. 2013;217(1):27-34. doi: 10.1016/j.jamcollsurg.2013.03.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.National Cancer Institute . Relative survival rate. Accessed September 28, 2022. https://www.cancer.gov/publications/dictionaries/cancer-terms/def/relative-survival-rate
- 24.Arias E, Heron M, Xu J. United States life tables, 2012. Natl Vital Stat Rep. 2016;65(8):1-65. [PubMed] [Google Scholar]
- 25.National Cancer Institute . Expected survival life tables—SEER documentation. Accessed March 8, 2023. https://seer.cancer.gov/expsurvival/index.html
- 26.Green AC, Baade P, Coory M, Aitken JF, Smithers M. Population-based 20-year survival among people diagnosed with thin melanomas in Queensland, Australia. J Clin Oncol. 2012;30(13):1462-1467. doi: 10.1200/JCO.2011.38.8561 [DOI] [PubMed] [Google Scholar]
- 27.Johnson MM, Leachman SA, Aspinwall LG, et al. Skin cancer screening: recommendations for data-driven screening guidelines and a review of the US Preventive Services Task Force controversy. Melanoma Manag. 2017;4(1):13-37. doi: 10.2217/mmt-2016-0022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Nosrati A, Yu WY, McGuire J, et al. Outcomes and risk factors in patients with multiple primary melanomas. J Invest Dermatol. 2019;139(1):195-201. doi: 10.1016/j.jid.2018.07.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Bradford PT, Goldstein AM, McMaster ML, Tucker MA. Acral lentiginous melanoma: incidence and survival patterns in the United States, 1986-2005. Arch Dermatol. 2009;145(4):427-434. doi: 10.1001/archdermatol.2008.609 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Bell KJL, Mehta Y, Turner RM, et al. Fear of new or recurrent melanoma after treatment for localised melanoma. Psychooncology. 2017;26(11):1784-1791. doi: 10.1002/pon.4366 [DOI] [PubMed] [Google Scholar]
- 31.Pomerantz H, Huang D, Weinstock MA. Risk of subsequent melanoma after melanoma in situ and invasive melanoma: a population-based study from 1973 to 2011. J Am Acad Dermatol. 2015;72(5):794-800. doi: 10.1016/j.jaad.2015.02.006 [DOI] [PubMed] [Google Scholar]
- 32.Trotter SC, Sroa N, Winkelmann RR, Olencki T, Bechtel M. A global review of melanoma follow-up guidelines. J Clin Aesthet Dermatol. 2013;6(9):18-26. [PMC free article] [PubMed] [Google Scholar]
- 33.Swetter SM, Tsao H, Bichakjian CK, et al. Guidelines of care for the management of primary cutaneous melanoma. J Am Acad Dermatol. 2019;80(1):208-250. doi: 10.1016/j.jaad.2018.08.055 [DOI] [PubMed] [Google Scholar]
- 34.Marsden JR, Newton-Bishop JA, Burrows L, et al. ; British Association of Dermatologists Clinical Standards Unit . Revised U.K. guidelines for the management of cutaneous melanoma 2010. Br J Dermatol. 2010;163(2):238-256. doi: 10.1111/j.1365-2133.2010.09883.x [DOI] [PubMed] [Google Scholar]
- 35.Swetter SM, Thompson JA, Albertini MR, et al. NCCN guidelines insights: melanoma: cutaneous, version 2.2021. J Natl Compr Canc Netw. 2021;19(4):364-376. doi: 10.6004/jnccn.2021.0018 [DOI] [PubMed] [Google Scholar]
- 36.Pflugfelder A, Kochs C, Blum A, et al. ; German Society of Dermatology . S3-guideline “diagnosis, therapy and follow-up of melanoma”–short version. J Dtsch Dermatol Ges. 2013;11(6):563-602. doi: 10.1111/ddg.12044 [DOI] [PubMed] [Google Scholar]
- 37.Cancer Council Australia . Ideal settings, duration and frequency of follow-up for patients with melanoma—Cancer Guidelines Wiki. Accessed December 28, 2022. https://wiki.cancer.org.au/australia/Clinical_question:What_is_the_ideal_setting,_duration_and_frequency_of_follow-up_for_melanoma_patients%3F#Follow-up_duration_and_frequency
- 38.Cullison CR, Zheng DX, Levoska MA, Bordeaux JS, Scott JF. Inconsistencies in cutaneous melanoma staging within SEER registries. JAMA Dermatol. 2021;157(6):727-729. doi: 10.1001/jamadermatol.2021.1098 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Elmore JG, Barnhill RL, Elder DE, et al. Pathologists’ diagnosis of invasive melanoma and melanocytic proliferations: observer accuracy and reproducibility study. BMJ. 2017;357:j2813. doi: 10.1136/bmj.j2813 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eMethods. Identification of cases of melanoma in situ
eFigure 1. Cohort flow diagram
eTable 1. Relative and disease-specific survival following melanoma in situ
eTable 2. Sensitivity analysis: Standardized mortality ratios for melanoma and all-causes following a diagnosis of a first and only melanoma in situ, by age at diagnosis, restricted to non-Hispanic White persons
eTable 3. Characteristics of patients who experienced a first primary MIS, by type of second primary
eFigure 2. Cumulative incidence of second primary melanoma among patients with a prior first primary melanoma in situ, by type of second primary
eTable 4. Standardized mortality ratios for disease-specific and all-cause mortality following a diagnosis of melanoma in situ and subsequent primary melanomas, by age at diagnosis
eFigure 3. Melanoma-specific survival following MIS by type of second primary melanoma
Data sharing statement