This observational cohort study assesses the incidence of new cutaneous melanoma in patients treated with ipilimumab, nivolumab, and/or pembrolizumab for metastatic melanoma.
Key Points
Question
What is the incidence of new primary cutaneous melanoma (CM) after initiating immune checkpoint inhibitor therapy for metastatic melanoma?
Findings
In this single-center observational cohort study, 42 of 2251 patients (1.9%) who received immune checkpoint inhibitors were diagnosed with 48 melanomas. The incidence rate was approximately 1100 cases per 100 000 person-years, and the cumulative cause-specific risk of new CM after 5 years was 4.9%; patients diagnosed with a new primary CM were more likely to have a family history of melanoma.
Meaning
Patients who receive checkpoint inhibitors for treatment of metastatic melanoma remain at risk for the development of new primary CMs.
Abstract
Importance
The development of new primary cutaneous melanoma (CM) after starting immune checkpoint inhibitor (ICI) therapy is poorly characterized.
Objective
To determine the incidence of new CM in patients treated with ipilimumab, nivolumab, and/or pembrolizumab for metastatic melanoma.
Design, Setting, and Participants
Single-center, retrospective, observational cohort study using an institutional database to identify patients diagnosed with melanoma at a tertiary care cancer hospital in New York, New York.
Exposures
Ipilimumab, nivolumab, and/or pembrolizumab treatment for metastatic melanoma.
Main Outcomes and Measures
Primary outcomes were the incidence proportion, the incidence rate, and the 5-year cause-specific cumulative risk.
Results
A total of 2251 patients were included in the study; mean (SD) age at the time of ICI start was 62.8 (14.4) years. The majority were male (63.8%, n = 1437), White (92.7%, n = 2086), and non-Hispanic (92.1%, n = 2073). Forty-two of 2251 patients who received ipilimumab, nivolumab, and/or pembrolizumab were diagnosed with 48 new CMs at a median (range) of 397.5 (39-2409) days after ICI initiation. The median age of affected patients at the time of ICI first dose was 66.5 years. The majority were male (66.7%, n = 28), White (92.9%, n = 39), and non-Hispanic (100.0%, n = 42). There were no differences in age, sex, race, and ethnicity among patients who did and did not develop a new CM. Patients who developed a new CM were more likely to have a family history of melanoma (23.8% vs 16.3%, P = .02). Most new CMs (n = 30, 62.5%) were diagnosed after the last date of ICI administration. Twenty-seven (56.3%) new CMs were in situ and 21 (43.8%) were invasive. Of the invasive CMs with a reported Breslow thickness (n = 20), the median (range) thickness was 0.4 (0.1-8.4) mm. The overall incidence proportion of new CM was 1.9% (95% CI, 1.4%-2.5%) and the incidence rate was 1103 cases per 100 000 person-years (95% CI, 815-1492). The 5-year cumulative cause-specific risk of new CM was 4.9% (95% CI, 3.3%-7.4%).
Conclusions and Relevance
Patients treated with ICI therapy for metastatic melanoma remain at risk for the development of new CM.
Introduction
Immune checkpoint inhibitors (ICIs) have been reported to lead to the resolution of unresectable primary cutaneous melanoma (CM), suggesting the potential role for an antitumor effect on synchronous or metachronous CM.1 A single-center study of 75 patients with metastatic melanoma who responded to systemic treatment identified only 1 patient who developed a new CM.2 Based on this observation, the authors questioned the utility of routine skin examinations in this patient population. To better inform surveillance guidelines, our objective was to identify the incidence of new CM in patients with metastatic melanoma who received ICIs.
Methods
After institutional review board approval was obtained, we searched a prospectively maintained institutional database of all patients diagnosed with melanoma. The need for informed consent was waived by the institutional review board because the research (1) involved no more than minimal risk to the participants or their privacy and (2) could not be practicably carried out without the waiver. From those who had ever received a US Food and Drug Administration–approved ICI for melanoma treatment (ie, ipilimumab, nivolumab, and/or pembrolizumab), we identified patients diagnosed with a new primary CM after the date of first administration of an ICI. The date of review cutoff was November 6, 2019. Definitions of variables are included in the eMethods in the Supplement.
Poisson regression models were used to estimate measures of incidence. If a patient developed more than 1 new CM, only the first new CM contributed to the analysis. Incidence proportion was estimated as the number of patients diagnosed with a new CM divided by the total number of patients at baseline who received ICI treatment. Incidence rate was estimated as the number of new CM diagnosed per 12 months of follow-up per 100 000 individuals. We also performed truncated incidence analyses in which the first 6 months of follow-up from all individuals (with or without a diagnosis of new CM) were removed. Cumulative cause-specific hazard estimates [H(t)] for new CM, along with 95% CIs, were estimated using the Nelson-Aalen estimator. A 2-sided P value less than .05 was considered statistically significant. Data analysis was performed with Stata, version 16.1 (StataCorp) from December 2019 to May 2020.
Results
A total of 2251 patients with metastatic melanoma who received ipilimumab, pembrolizumab, and/or nivolumab between 2003 and 2019 were identified. The mean (SD) age of patients at the time of ICI start was 62.8 (14.4) years. The majority were male (63.8%, n = 1437), White (92.7%, n = 2086), and non-Hispanic (92.1%, n = 2073) (Table 1). Over the course of treatment, patients most frequently received ipilimumab alone (n = 793, 35.2%); ipilimumab and nivolumab (n = 543, 24.1%); pembrolizumab alone (n = 330, 14.7%); ipilimumab and pembrolizumab (n = 227, 10.1%); ipilimumab, nivolumab, and pembrolizumab (n = 211, 9.4%); nivolumab alone (n = 137, 6.1%); or nivolumab and pembrolizumab (n = 10, 0.4%). Median duration of time from first dose of ICI to the first new CM, last follow-up, or death was 12.1 months. A total of 724 patients (32.2%) were seen by dermatology at our institution for an outpatient visit for any reason after starting ICI therapy. At the time of last follow-up or contact, 46.2% (n = 1041) of patients were alive and 53.8% (n = 1210) were deceased.
Table 1. Demographic Characteristics of Patients Who Received Immune Checkpoint Inhibitors for Metastatic Melanoma.
| Variable | No. (%) | P value | ||
|---|---|---|---|---|
| New CM | Total (n = 2251) | |||
| No (n = 2209) | Yes (n = 42) | |||
| Age, mean (SD), y | 62.7 (14.4) | 66.6 (12.4) | 62.8 (14.4) | .08a |
| Sex | ||||
| Female | 800 (36.2) | 14 (33.3) | 814 (36.2) | .70b |
| Male | 1409 (63.8) | 28 (66.7) | 1437 (63.8) | |
| Race | ||||
| Asian/Indian | 40 (1.8) | 0 | 40 (1.8) | .53c |
| Black | 30 (1.4) | 0 | 30 (1.3) | |
| Native American | 1 (0) | 0 | 1 (0) | |
| White | 2047 (92.7) | 39 (92.9) | 2086 (92.7) | |
| Other | 22 (1.0) | 0 | 22 (1.0) | |
| Patient refused | 62 (2.8) | 3 (7.1) | 65 (2.9) | |
| Unknown | 7 (0.3) | 0 | 7 (0.3) | |
| Ethnicity | ||||
| Hispanic | 44 (2.0) | 0 | 44 (2.0) | .23c |
| Not Hispanic | 2031 (91.9) | 42 (100.0) | 2073 (92.1) | |
| Unknown | 134 (6.1) | 0 | 134 (6.0) | |
| Family history of melanoma | ||||
| No | 1804 (81.7) | 30 (71.4) | 1834 (81.5) | .02b |
| Yes | 360 (16.3) | 10 (23.8) | 370 (16.4) | |
| Unknown | 45 (2.0) | 2 (4.8) | 47 (2.1) | |
Abbreviation: CM, cutaneous melanoma.
P value based on t test.
P value based on χ2 statistic.
P value based on Fisher exact test.
Forty-two patients with 48 new CMs were identified (Table 2). The median age of affected patients at the time of ICI first dose was 66.5 years. The majority were male (66.7%, n = 28), White (92.9%, n = 39), and non-Hispanic (100.0%, n = 42). There were no differences in age, sex, race, and ethnicity among patients who did and did not develop a new CM. Patients with a new CM were more likely to have a family history of melanoma than patients without a new CM (23.8% vs 16.3%, P = .02). Eighteen new CMs (37.5%) were diagnosed before the last date of ICI administration, and 30 (62.5%) were diagnosed after the last date. The median (range) time from the first ICI dose to the diagnosis of each patient’s first new CM was 397.5 (39-2409) days. Twenty-seven (56.3%) melanomas were in situ, and 21 (43.8%) were invasive. Of the invasive melanomas with a reported Breslow thickness (n = 20), the median (range) thickness was 0.4 (0.1-8.4) mm. The first new CM was detected by a physician in 26 cases (61.9%), the patient/family in 10 cases (23.8%), and unknown in 6 cases (14.3%). The first new CM was identified through a screening examination in 17 patients (40.5%). The systemic treatments that patients received before and after ICI start are detailed in Table 2 and eResults in the Supplement; of note, 6 patients (14.3%) had received vemurafenib monotherapy prior to the diagnosis of new CM (2 before ICI start; 4 after ICI start and before new CM). At the time of last follow-up or contact, 29 (69.0%) patients were alive and 13 (31.0%) were dead.
Table 2. Characteristics of Patients With New Cutaneous Melanoma After Starting Immune Checkpoint Inhibitors.
| Patient No. | AJCC (8th ed) stage at ICI start (unknown primary) | Primary CMs before ICI start, No. | ICI received before new primary CM diagnosis | ICI start to new primary CM diagnosis, d | Anatomic site (Breslow thickness) of new primary CMs diagnosed after ICI start | ICI received after new primary CM(s) diagnosed | ICI duration, first to last dose, d | Other treatment of metastatic disease | ICI start to last follow-up or death, d | Vital/disease status | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Before ICI start | After ICI start | ||||||||||
| 1 | IV (no) | 1 | Ipi | 404 | Trunk (in situ)a,b | NA | 84 | S | NA | 2709 | Alive, NED |
| 2 | IV (no) | 2 | Pembro | 508 | Trunk (2.5 mm)c,d | NA | 189 | R, S | S | 977 | Alive, NED |
| 3 | IV (no) | 3 | Ipi | 391 | Trunk (1.1 mm)a,b | Pembro, nivo | 2331 | R, S | S, C | 2331 | Alive, NED |
| 4 | IV (no) | 2 | Ipi | 162 | Trunk (in situ)a,d | Ipi, nivo, pembro | 2029 | NA | R, S, T, I | 2139 | Alive, WD |
| 5 | IV (no) | 3 | Ipi | 63 | H&N (3 mm)c,d | NA | 25 | R, S | R | 2274 | Alive, NED |
| 340 | H&N (0.3 mm) | ||||||||||
| 869 | Trunk (in situ) | ||||||||||
| 6 | IV (no) | 1 | Ipi | 1347 | UE (in situ)a,e | NA | 73 | S | NA | 2853 | Alive, NED |
| 7 | III (no) | 6 | Ipi | 73 | UE (0.4 mm)c,d | NA | 31 | NA | C | 2362 | Alive, NED |
| 8 | IV (no) | 1 | Ipi | 593 | Trunk (in situ)a,b | NA | 42 | S, C, T | NA | 3658 | Alive, NED |
| 9 | IV (no) | 7 | Pembro | 531 | LE (0.3 mm)a,d | NA | 341 | NA | NA | 611 | Alive, NED |
| 10 | IV (no) | 1 | Ipi, nivo | 492 | H&N (in situ)e,f | NA | 42 | R, S, V | NA | 2597 | Alive, NED |
| 11 | IV (yes) | 1 | Ipi | 240 | UE (0.3 mm)a,d | NA | 583 | S, C | S | 3590 | Alive, NED |
| 12 | IV (yes) | 0 | Ipi | 280 | H&N (0.62 mm)c,d | NA | 42 | NA | R, C, T | 492 | Dead, DOD |
| 13 | IV (no) | 1 | Ipi, Nivo | 365 | Trunk (0.45 mm)a,b | Ipi, nivo | 966 | S, I | R, S, T | 1486 | Dead, DOD |
| 14 | IV (no) | 1 | Ipi | 66 | UE (0.3 mm)c,d | Ipi | 476 | S, V | R, T | 519 | Dead, DOD |
| 15 | IV (no) | 1 | Ipi | 338 | LE (invasive)c,d | Ipi | 735 | S | R, C, T | 759 | Dead, DOD |
| 16 | IV (no) | 2 | Pembro | 765 | H&N (in situ)a,b | NA | 65 | NA | NA | 968 | Alive, NED |
| 17 | IV (no) | 2 | Ipi | 225 | LE (in situ)c,d | Ipi | 646 | C | C, T | 680 | Dead, DOD |
| 254 | H&N (0.3 mm) | ||||||||||
| 18 | IV (no) | 1 | Ipi | 1435 | UE (in situ)a,b | NA | 63 | S | NA | 1727 | Dead, DWD |
| 19 | IV (no) | 1 | Pembro | 259 | UE (0.2 mm)d,f | Ipi | 667 | R | R, I | 989 | Alive, WD |
| 20 | IV (no) | 1 | Ipi, nivo, pembro | 382 | H&N (in situ)a,b | NA | 962 | R, S | NA | 992 | Alive, NED |
| 21 | IV (yes) | 1 | Ipi, nivo, pembro | 1149 | Trunk (0.38 mm)a,d | NA | 849 | R, S | R | 1332 | Alive, WD |
| 22 | IV (yes) | 0 | Pembro | 592 | LE (in situ)a,b | NA | 732 | NA | NA | 1318 | Alive, NED |
| 23 | IV (yes) | 0 | Nivo, pembro | 488 | UE (in situ)c,d | NA | 239 | NA | NA | 693 | Alive, NED |
| 663 | LE (0.46 mm) | ||||||||||
| 24 | IV (no) | 2 | Ipi | 991 | Trunk (in situ)a,d | NA | 77 | S, V | NA | 2414 | Alive, NED |
| 25 | IV (no) | 1 | Pembro | 159 | H&N (in situ)c,d | NA | 33 | NA | R, S | 505 | Dead, DWD |
| 26 | III (no) | 1 | Ipi | 39 | LE (in situ)a,d | NA | 42 | S, T | NA | 199 | Dead, DOD |
| 27 | IV (no) | 5 | Ipi, pembro | 2409 | H&N (in situ)a,d | NA | 2277 | S, C | NA | 3094 | Alive, NED |
| 28 | IV (no) | 3 | Ipi, nivo | 391 | Trunk (0.54 mm)a,b | NA | 140 | S | NA | 1680 | Alive, NED |
| 29 | III (yes) | 2 | Ipi, nivo | 531 | H&N (0.15 mm)a,b | NA | 42 | NA | NA | 2490 | Alive, NED |
| 30 | IV (yes) | 1 | Ipi | 1477 | H&N (in situ)e,f | NA | 274 | NA | R | 2894 | Dead, NOS |
| 31 | IV (no) | 1 | Ipi | 120 | H&N (in situ)e,f | NA | 66 | R, S | NA | 148 | Dead, DOD |
| 32 | III (no) | 1 | Ipi | 203 | LE (0.4 mm)a,b | Pembro | 322 | NA | S, C | 1467 | Dead, DOOC |
| 254 | H&N (in situ) | ||||||||||
| 33 | IV (no) | 2 | Ipi, nivo, pembro | 693 | Trunk (0.5 mm)a,b | Pembro | 1232 | NA | R, S, C | 1308 | Dead, DOD |
| 34 | IV (no) | 6 | Ipi, pembro | 1740 | Trunk (in situ)a,b | NA | 1218 | R, S, C, T, I | S, T | 2769 | Alive, NED |
| 2769 | UE (in situ) | ||||||||||
| 35 | IV (no) | 2 | Ipi | 2140 | H&N (8.4 mm)e,f | Ipi | 2588 | S | S | 2631 | Alive, WD |
| 36 | III (no) | 1 | Ipi | 844 | Trunk (in situ)a,b | NA | 62 | S, C, I, V, GM-CSF, TNF, IL DNCB | NA | 3518 | Alive, NED |
| 37 | III (no) | 1 | Ipi | 2160 | LE (in situ)c,d | NA | 522 | S | S, C | 2862 | Alive, NED |
| 38 | IV (yes) | 0 | Ipi | 78 | Trunk (in situ)a,d | NA | 64 | S | S | 1800 | Alive, NED |
| 39 | IV (yes) | 1 | Pembro | 118 | Trunk (in situ)a,b | NA | 727 | S | R | 902 | Alive, NED |
| 40 | IV (no) | 1 | Ipi, pembro | 351 | Trunk (in situ)a,b | NA | 125 | NA | R | 1703 | Alive, WD |
| 41 | IV (yes) | 0 | Ipi | 699 | UE (0.14 mm)f,e | NA | 62 | R, S | R | 1806 | Alive, WD |
| 42 | IV (no) | 1 | Ipi, nivo, pembro | 182 | Trunk (in situ)a,b | Nivo | 336 | S | C, T, TVEC | 542 | Dead, DOD |
Abbreviations: AJCC, American Joint Committee on Cancer; C, cytotoxic chemotherapy; CM, cutaneous melanoma; DOD, died of melanoma; DOOC, died of other cause; DWD, died with active disease but of unknown cause; GM-CSF, granulocyte-macrophage colony-stimulating factor; H&N, head and neck; I, other immunotherapy; ICI, immune checkpoint inhibitor; IL DNCB, intralesional dinitrochlorobenzene; ipi, ipilimumab; LE, lower extremity; NA, not applicable; NED, no evidence of disease; nivo, nivolumab; NOS, not otherwise specified; pembro, pembrolizumab; R, radiation; S, surgery; T, VEGFR/PDGFR/multikinase/RAF/BRAF/MEK/PI3K-γ/ERK/CDK targeted therapy; TNF, tumor necrosis factor; TVEC, talimogene laherparepvec; UE, upper extremity; V, vaccine; WD, with disease.
First identified by physician.
Identified during screening examination.
First identified by patient or family.
Identified during other visit type.
Unknown if identified during screening examination or other visit type.
Unknown who first identified lesion.
The incidence proportion of new CM was 1.9% (95% CI, 1.4%-2.5%), and the incidence rate was 1103 cases per 100 000 person-years (95% CI, 815-1492). In the 1546 individuals who had 6 or more months of uncensored follow-up time, the incidence proportion was 2.1% (95% CI, 1.5%-3.0%), and the rate was 1144 cases per 100 000 person-years (95% CI, 813-1608). The cumulative cause-specific hazard of new CM is shown in the Figure and shows that the incidence rate was approximately constant over time. The cumulative cause-specific 2-year and 5-year risks of new CM were 2.6% (95% CI, 1.8%-3.7%) and 4.9% (95% CI, 3.3%-7.4%), respectively.
Figure. Cumulative Hazard of New Primary Cutaneous Melanoma After Starting Immune Checkpoint Inhibitor Treatment.
The cause-specific hazard is the instantaneous rate of developing the event (ie, new primary cutaneous melanoma) at an individual time point (t), given that the individual has survived up to time t. The hazard is not a probability and does not have an upper bound. Specifically, the cumulative hazard should not be interpreted as the cumulative risk of new primary cutaneous melanoma. Rather, it provides a visual depiction of the relationship between the change in the rate of developing a new melanoma and time after immune checkpoint inhibitor initiation. This figure shows that the rate of new melanoma is approximately constant over time.
Discussion
In a single-institution cohort of more than 2000 patients with metastatic melanoma treated with ICIs, the incidence proportion of new CM was 1.9%, the 5-year cumulative risk was 4.9%, and the incidence rate was 1103 cases per 100 000 person-years. DiFronzo et al3 found that the incidence proportion and rate of a second primary melanoma among patients with a history of localized CM (n = 3310) were 3.4% and 325 per 100 000 person-years, respectively. An analysis of 977 Greek patients with melanoma4 found the incidence proportion of second primary melanoma to be 4.7% and the 5-year cause-specific cumulative risk to be 8%. Comparisons with these cohorts are challenging owing to differences in geography, health systems, time periods, and patient populations. A previous study at our institution5 examined 4484 patients with a first primary melanoma diagnosed from 1996 through 2002 and found that the incidence proportion of multiple primary melanoma was 8.6%. In a competing-risk incidence analysis, the cumulative 5-year risk of a second primary melanoma was 11.4%, more than double the risk found in our study. These historical data from our institution suggest the possibility that ICI treatment could have reduced the risk of the development of a new CM; however, given that our study is observational, we view these findings as hypothesis generating.
The new CMs diagnosed in our cohort may have developed prior to, during, or subsequent to ICI treatment. Five CMs were diagnosed within 90 days of drug initiation, which strongly points to these being preexisting tumors. Of note, 14 patients did not receive any subsequent systemic treatment modalities after ICI therapy and were alive without evidence of metastatic disease, suggesting that they were diagnosed with new CMs despite a likely therapeutic response to ICI.
Limitations
This study is limited by its single-center, retrospective, observational design. We are unaware of the proportion of patients who received a skin cancer screening examination at our or another institution; however, approximately a third of patients who received ICI treatment saw a dermatologist at our institution for an outpatient visit of any reason, including an ICI-related skin adverse event. More than a third of patients received ipilimumab alone, which is no longer a first-line treatment for metastatic melanoma. Finally, a subset of patients also received vemurafenib monotherapy, which may have predisposed them to new CM development.6
Conclusions
Patients with metastatic melanoma treated with ICIs remain at risk for the development of new CMs. Prospective studies with larger sample sizes and longer follow-up are needed to better assess risk factors for the development of new CM following ICI treatment as well as any potential benefits from screening patients who respond to therapy.
eMethods.
eResults.
References
- 1.Navarrete-Dechent C, Cordova M, Postow MA, et al. Evaluation of the response of unresectable primary cutaneous melanoma to immunotherapy visualized with reflectance confocal microscopy: a report of 2 cases. JAMA Dermatol. 2019;155(3):347-352. doi: 10.1001/jamadermatol.2018.3688 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Huis In’t Veld EA, van Seventer IC, van Teeseling L, van Thienen JV, Crijns MB. Is routine skin examination in metastatic melanoma patients treated with immuno- or targeted therapy necessary? J Eur Acad Dermatol Venereol. 2020;34(4):e175-e176. doi: 10.1111/jdv.16128 [DOI] [PubMed] [Google Scholar]
- 3.DiFronzo LA, Wanek LA, Elashoff R, Morton DL. Increased incidence of second primary melanoma in patients with a previous cutaneous melanoma. Ann Surg Oncol. 1999;6(7):705-711. doi: 10.1007/s10434-999-0705-0 [DOI] [PubMed] [Google Scholar]
- 4.Lallas A, Apalla Z, Kyrgidis A, et al. Second primary melanomas in a cohort of 977 melanoma patients within the first 5 years of monitoring. J Am Acad Dermatol. 2020;82(2):398-406. doi: 10.1016/j.jaad.2019.08.074 [DOI] [PubMed] [Google Scholar]
- 5.Ferrone CR, Ben Porat L, Panageas KS, et al. Clinicopathological features of and risk factors for multiple primary melanomas. JAMA. 2005;294(13):1647-1654. doi: 10.1001/jama.294.13.1647 [DOI] [PubMed] [Google Scholar]
- 6.Dalle S, Poulalhon N, Debarbieux S, et al. Tracking of second primary melanomas in vemurafenib-treated patients. JAMA Dermatol. 2013;149(4):488-490. doi: 10.1001/jamadermatol.2013.21 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eMethods.
eResults.