Key Points
Question
What is the association between cutaneous immune-related adverse events (cirAEs) and noncutaneous immune-related adverse events (irAEs) among patients with cirAEs associated with immune checkpoint-inhibitor therapy?
Findings
In this cohort study of 358 patients with cancer, the first cirAEs often preceded subsequent irAEs. In addition, associations were observed between several types of cirAEs and types of irAEs.
Meaning
Patients with certain subtypes of early cirAEs may be at elevated risk for developing irAEs, suggesting a potential role for dermatologists in the management of these events.
Abstract
Importance
Cutaneous immune-related adverse events (cirAEs) are some of the earliest toxic reactions to emerge following immune-checkpoint inhibitor (ICI) initiation. As an early indicator of robust inflammatory response, cirAEs may be associated with patterns of immune-mediated toxic effects, but associations between these events and noncutaneous immune-related adverse events (irAEs) remain underexplored.
Objectives
To characterize patterns of cirAEs and irAEs across care settings and examine associations between the features of first cirAE, overall irAE risk, and risk of specific irAE subtypes.
Design, Setting, and Participants
A retrospective cohort study was conducted at a single academic medical center. The cohort included 358 patients with cancer who initiated anti–programmed death 1/ligand 1 and/or anticytotoxic-T-lymphocyte-4 ICI therapy between January 1, 2016, and March 8, 2019, and developed 1 or more cirAEs, identified using International Statistical Classification of Diseases and Related Health Problems, Tenth Revision codes and confirmed via manual medical record review. All relevant information documented before March 31, 2020, was included.
Exposures
Anti–programmed death 1/ligand 1 and/or anticytotoxic-T-lymphocyte-4 therapy.
Main Outcomes and Measures
Associations between specific cirAE morphologic classes and patterns of irAEs (occurrence, timeline, organ class, and specific toxic effects). Given the potential that shared underlying factors are associated with the risk of both noncutaneous and cutaneous toxic effects, the presence of observed positive associations between certain cirAE and irAE subtypes was hypothesized.
Results
Of the 358 patients, 213 were men (59.5%); median age was 65 years (interquartile range, 55-73 years). Nearly half of the patients (177 [49.4%]) with cirAE also developed a noncutaneous irAE. Most patients (128 [72.3%]) experienced their first cirAE before developing any irAE. Several cirAE morphologic classes were found to be associated with overall, organ-based, and specific irAEs. More specifically, mucositis was found to be associated with overall irAE risk (odds ratio [OR], 5.28; 95% CI, 1.11-24.26; P = .04), gastrointestinal irAEs (OR, 5.70; 95% CI, 1.11-29.40; P = .04), and the specific diagnosis of gastroenterocolitis (OR, 6.80; 95% CI, 1.24-37.39; P = .03). In addition, psoriasis was associated with an increased risk of endocrine irAEs (OR, 4.54; 95% CI, 1.21-17.04; P = .03).
Conclusions and Relevance
In this cohort study, these findings underscore the risk of multisystem toxic effects in patients experiencing cirAEs and highlight potential opportunities for dermatologists in the management of noncutaneous toxic effects.
This cohort study examines the incidence of noncutaneous immune-related adverse events in patients with cancer who have developed cutaneous immune-related adverse events during therapy with immune-checkpoint inhibitors.
Introduction
Cutaneous immune-related adverse events (cirAEs) are some of the earliest toxic effects to emerge following initiation of immune-checkpoint inhibitor (ICI) therapy.1,2,3 Associations between cirAEs and noncutaneous immune-related (irAEs) remain underexplored.4,5,6,7,8,9,10,11,12,13 In this retrospective cohort study, we sought to address this lack of information by characterizing patterns of cirAE and irAE concomitant occurrence across care settings and examining associations between the features of the first cirAE, overall irAE risk, and risk of specific irAE subtypes.
Methods
Participant Characteristics
We screened 2459 patients aged 18 years and older who initiated anti–programmed death 1/ligand 1 (PD-1/PDL-1) and/or anticytotoxic-T-lymphocyte-4 (CTLA-4) ICI therapy at Massachusetts General Hospital between January 1, 2016, and March 16, 2019. The Massachusetts General Brigham Institutional Review Board approved this study.
Screening for possible cirAEs was performed using International Statistical Classification of Diseases and Related Health Problems, Tenth Revision codes (n = 910) (eTable 1 in the Supplement), with subsequent confirmation of cirAE status through manual review of the medical records.1,3,9 Eligible cirAEs included reactions emerging after ICI initiation consistent with established morphologic categories, involving greater than 1% of the body surface area, lasting more than 1 day, and attributed to ICIs by the evaluating clinician. Cases with borderline features were reviewed by 2 or more study personnel, including at least 1 board-certified dermatologist (S.T.C.).1,3,9
For each patient with cirAEs (n = 358), we abstracted demographic characteristics, pre-ICI treatments (chemotherapy, targeted agents, other immunotherapy, and radiotherapy) and oncologic history for the date of ICI therapy initiation. Targeted agents were defined as antineoplastic therapies targeting specific tumor-associated molecules (ie, vascular endothelial growth factor). Other immunotherapies were defined as non-ICI antineoplastic agents amplifying the host antitumor response (eg, interferon). We abstracted cirAE features using dermatologist characterizations of cirAEs when available2 and specific diagnoses, grouping them into 7 organ-based classes: (1) endocrine, (2) gastrointestinal, (3) hepatobiliary, (4) pulmonary, (5) rheumatologic, (6) ocular, and (7) other.
Statistical Analysis
Associations between features of the first cirAE and the risk of any irAE and associations between features of the first cirAE and irAE subtypes were assessed using logistic regression. Logistic regression models were created for all irAE organ classes with a frequency greater than 20, and for specific irAE diagnoses with a frequency greater than 20 differing from organ class category by more than 5 patients. Three specific irAE diagnoses fit this criterion: thyroiditis, gastroenterocolitis, and arthritis. All logistic regression models were adjusted for age, sex, ICI regimen, and covariates, with a 2-sided P value <.05 considered significant. To address multiple hypothesis testing, we used the false discovery rate control method with a false discovery rate of 0.15.14
Results
A total of 358 of the screened patients developed cirAEs (14.5% of all patients who initiated ICI therapy). Median age was 65 years (interquartile range [IQR], 55-73 years), 213 patients (59.5%) were men, 145 (40.5%) were women, and 150 patients (41.9%) had melanoma. Initial cirAEs emerged a median of 49 days (IQR, 19-146 days) after ICI initiation, with a median peak Common Terminology Criteria for Adverse Events, Version 5.0, grade of 1 (IQR, 1-2), and most often included maculopapular or papulopustular reactions (186 [52.0%]), pruritus without visible cutaneous manifestations (isolated pruritus, 66 [18.4%]), and eczematous reactions (25 [7.0%]) (Table 1; eTable 1 in the Supplement). Eighty-two patients (22.9%) developed multiple cirAEs.
Table 1. Associations Between Patient Characteristics, cirAE Features, and Emergence of Any irAE in 358 Patients.
| Characteristic | No. (%) | OR (95% CI) | P valuea | ||
|---|---|---|---|---|---|
| Total cirAE patients (n = 358) | Patients | ||||
| With irAE (n = 177) | Without irAE (n = 181) | ||||
| Age, median (IQR), y | 65 (55-73) | 64 (54-72) | 66 (56-74) | NS | NS |
| Sex | |||||
| Women | 145 (40.5) | 72 (40.7) | 73 (59.7) | NS | NS |
| Men | 213 (59.5) | 105 (49.3) | 108 (50.7) | ||
| Cancer type | |||||
| Melanoma | 150 (41.9) | 113 (63.8) | 37 (20.4) | [Reference] | |
| Lung | 59 (16.5) | 18 (10.2) | 41 (22.7) | 0.15 (0.08-0.32) | <.001 |
| Head or neck | 32 (8.9) | 10 (5.6) | 22 (12.2) | 0.17 (0.07-0.44) | <.001 |
| Gastrointestinal | 34 (9.5) | 12 (6.8) | 22 (12.2) | 0.18 (0.08-0.47) | <.00 |
| Otherb | 83 (23.2) | 24 (13.6) | 59 (32.6) | 0.17 (0.09-0.34) | <.001 |
| Cancer stage, median (IQR) | 4 (3-4) | 4 (3-4) | 4 (3-4) | NS | NS |
| ECOG PMR, median (IQR) | 1 (1-2) | 1 (1-2) | 1 (1-2) | NS | NS |
| Pre-ICI treatments | |||||
| Radiotherapy | 141 (39.4) | 53 (29.9) | 88 (48.6) | NS | NS |
| Traditional chemotherapy | 150 (41.9) | 45 (25.4) | 105 (58.0) | NS | NS |
| Targeted agent | 67 (18.7) | 18 (10.2) | 49 (27.1) | 0.33 (0.16-0.69) | .003 |
| Other immunotherapy | 19 (5.3) | 13 (7.3) | 6 (3.3) | NS | NS |
| ICI precipitating first cirAE or irAE | |||||
| Anti–PD-1/PDL-1 | 275 (76.8) | 119 (67.2) | 156 (86.2) | 1 [Reference] | |
| Anti–CTLA-4 | 14 (3.0) | 13 (7.3) | 1 (0.6) | NS | |
| Anti–PD-1/PDL-1 + CTLA-4 | 69 (19.3) | 45 (25.6) | 24 (13.3) | 2.78 (1.42-5.44) | .003 |
| Time to first cirAE, median (IQR), dc,d | 49 (19-146) | 43 (17-126) | 53 (20-166) | NS | NS |
| Morphologic characteristics of first cirAE | |||||
| Maculopapular or papulopustular, NOS | 186 (52.0) | 96 (54.2) | 90 (49.7) | 1 [Reference] | |
| Drug hypersensitivity, NOS | 17 (4.7) | 9 (5.1) | 8 (4.4) | NS | NS |
| Eczematous | 25 (7.0) | 9 (5.1) | 16 (8.8) | NS | NS |
| Lichenoid | 14 (3.9) | 5 (2.8) | 9 (5.0) | NS | NS |
| Isolated pruritus | 66 (18.4) | 32 (18.1) | 34 (18.8) | NS | NS |
| Psoriasis | 12 (3.4) | 7 (4.0) | 5 (2.8) | NS | NS |
| Vitiligo | 9 (2.5) | 7 (4.0) | 2 (1.1) | NS | NS |
| Mucositis | 9 (2.5) | 6 (3.4) | 3 (1.7) | 5.28 (1.11-24.26) | .04 |
| Othere | 20 (5.6) | 6 (3.4) | 14 (7.7) | NS | NS |
| Multiple cirAEs | 82 (22.9) | 55 (31.1) | 27 (14.9) | NS | NS |
| Peak CTCAE severity of first cirAE, median (IQR) | 1 (1-2) | 1 (1-2) | 1 (1-2) | NS | NS |
| Multiple irAEs | 67 (37.9) | NA | NA | ||
| Time to first irAE, median (IQR), d | 96 (50-190) | NA | NA | ||
| Time to all irAEs, median (IQR), d | 115 (57-229) | NA | NA | ||
Abbreviations: cirAE, cutaneous immune-related adverse event; CTCAE, Common Terminology Criteria For Adverse Events, Version 5.0; CTLA-4, cytotoxic T-lymphocyte antigen-4; ECOG, Eastern Cooperative Oncology Group; ICI, immune checkpoint inhibitor; IQR, interquartile range; irAE, noncutaneous immune-related adverse event; NA, not applicable; NOS, not otherwise specified; NS, not significant; OR, odds ratio; PD-1, programmed cell death-1; PD-L1, PD-1-ligand 1; PMR, polymyalgia rheumatica.
A P value <.05 represents a significant association between the specific patient feature and occurrence of any irAE, using the stated reference category as baseline.
Other cancer types included genitourinary (renal, urothelial), gynecologic (cervical, ovarian), breast, hematologic (Hodgkin lymphoma, non-Hodgkin lymphoma), cutaneous (nonmelanoma skin cancers, Merkel cell carcinoma), endocrine (neuroendocrine, adrenocortical, thyroid), musculoskeletal/soft tissue (sarcoma), and neurologic (meningioma, glioblastoma) cancers.
Median time to onset of initial mucositis cirAE and first irAE, when co-occurring: 21 days (IQR, 17-41 days) and 231 days (IQR, 46-424 days). Median time to onset of initial mucositis cirAE and gastrointestinal irAE (including gastroenterocolitis), when co-occurring: 21 days (IQR, 16-21 days) and 63 days (IQR, 40-388 days). Median time to onset of initial mucositis cirAE and rheumatologic irAE (including arthritis), when co-occurring: 199 days (IQR, 41-357 days) and 416 days (IQR, 407-424 days).
Median time to onset of initial psoriasiform cirAE and endocrine irAE, when co-occurring: 181 days (IQR, 108-323 days) and 180 days (IQR, 154-181 days). Median time to onset of initial psoriasiform cirAE and thyroiditis irAE, when co-occurring: 252 days (IQR, 94-399 days) and 167 days (IQR, 150-180 days).
Other morphologic characteristics of first cirAE included bullous pemphigoid (overall number of patients, 7; with irAE, 0; without irAE, 7), urticarial reaction (overall, 4; with irAE, 1; without irAE, 3), erythema multiformelike reaction (overall, 4; with irAE, 3; without irAE, 1), panniculitis (overall, 2; with irAE, 1; without irAE, 1), Stevens-Johnson syndromelike reaction (overall, 1; with irAE, 0; without irAE, 1), Sweet syndrome (overall, 1; with irAE, 0; without irAE, 1), and dyshidrotic eczema (overall, 1; with irAE, 1; without irAE, 0).
A total of 177 patients (49.4%) with cirAEs also developed a noncutaneous irAE. Most of these patients (128 [72.3%]) experienced their first cirAE before developing any irAE and 67 patients (18.7%) experienced multiple irAEs, resulting in 274 total irAEs (Table 2). The most common irAEs by organ class were endocrine (84 [30.7%]), gastrointestinal (66 [24.1%]), and hepatobiliary (40 [14.6%]) (Table 2). The most common irAEs by specific diagnosis were thyroiditis (53 [19.3%]), gastroenterocolitis (57 [20.8%]), and hepatotoxicity (39 [14.2%]).
Table 2. Features of irAEs.
| Characteristic | No. (%) |
|---|---|
| Total irAEs | 274 |
| irAE type | |
| Endocrine | 84 (30.7) |
| Adrenal insufficiency | 12 (4.4) |
| Immune-mediated diabetes | 4 (1.5) |
| Thyroiditis | 53 (19.3) |
| Hypophysitis | 15 (5.5) |
| Gastrointestinal | 66 (24.1) |
| Gastroenterocolitis | 57 (20.8) |
| Pancreatitis | 7 (2.6) |
| Celiac | 1 (0.4) |
| Uvulitis | 1 (0.4) |
| Hepatobiliary | 40 (14.6) |
| Hepatotoxicity | 39 (14.2) |
| Cholangitis | 1 (0.4) |
| Pulmonary | 27 (9.9) |
| Pneumonitis | 27 (9.9) |
| Rheumatologic | 30 (10.9) |
| Arthritis | 23 (8.4) |
| PMR | 1 (0.4) |
| Myositis | 3 (8.4) |
| Sicca | 2 (0.7) |
| Vasculitis | 1 (0.4) |
| Ocular | 11 (4.0) |
| Uveitis | 6 (2.2) |
| Conjunctivitis | 5 (1.8) |
| Othera | 16 (5.8) |
Abbreviations: GI, gastrointestinal; IQR, interquartile range; irAE, noncutaneous immune-related adverse event; PMR, polymyalgia rheumatica.
Other irAEs included renal (nephritis, n = 6), neurologic (encephalitis, n = 2), peripheral neuropathy (n = 3), cardiac (myocarditis, n = 4), and systemic inflammatory response syndrome (n = 1).
With regard to the association between the first cirAE and likelihood of any irAE, patients with mucositis were significantly more likely to develop an irAE compared with those with other cirAE types (odds ratio [OR], 5.28; 95% CI, 1.11-24.26; P = .04). Considering associations between specific morphologic classes of the first cirAE and irAE risk by organ class or specific toxic effect, several significant associations were also observed between patterns of irAE and mucositis. Mucositis was found to be associated with gastrointestinal irAEs (OR, 5.70; 95% CI, 1.11-29.40; P = .04) and the specific diagnosis of gastroenterocolitis (OR, 6.80; 95% CI, 1.24-37.39; P = .03). Development of mucositis (median onset, 21 days) preceded the emergence of any irAE (median onset, 231 days), including emergence of gastrointestinal irAEs (median onset, 63 days).
Another subtype of cirAE, psoriasiform reactions, was also found to be significantly associated with irAE risk by organ class and specific toxic effect. More specifically, individuals with an initial cirAE of this subtype were found to have more frequent endocrine irAEs (OR, 4.54; 95% CI, 1.21-17.04; P = .03), with similar median time to onset (181 days for psoriasiform cirAEs vs 180 days for endocrine irAEs). The remaining irAE organ classes and specific diagnoses were either too infrequent or not significantly associated with any specific cirAE subtypes. Considering the severity of the first cirAE and irAEs, both by organ class and specific irAE type, no significant associations were observed (Table 3).
Table 3. Associations Between cirAE Features and irAE Types.
| Characteristic | OR (95% CI) | P value |
|---|---|---|
| Organ class of irAE | ||
| Endocrine irAE | ||
| cirAE morphologic characteristic (psoriasiform)b | 4.54 (1.21-17.04) | .03a |
| cirAEseverity | 0.83 (0.34-2.03) | .68 |
| GI irAE | ||
| cirAE morphologic characteristic (mucositis)b | 5.70 (1.11-29.40) | .04a |
| cirAEseverity | 1.74 (0.72-4.20) | .22 |
| Hepatobiliary irAE | ||
| cirAE morphologic characteristicc | NS | NS |
| cirAEseverity | 0.44 (0.10-1.98) | .28 |
| Pulmonary irAE | ||
| cirAE morphologic characteristicc | NS | NS |
| cirAEseverity | 1.3 (0.41-4.11) | .65 |
| Rheumatologic irAE | ||
| cirAE morphologic characteristic (mucositis)b | 7.87 (1.18-52.74) | .03e |
| cirAEseverity | 0.27 (0.03-2.07) | .21 |
| Specific irAE diagnosis | ||
| Thyroiditis | ||
| cirAE morphologic characteristic (psoriasiform)d | 3.68 (0.93-14.51) | .06f |
| cirAEseverity | 0.59 (0.19-1.80) | .35 |
| Gastroenterocolitis | ||
| cirAE morphologic characteristic (mucositis)d | 6.80 (1.24-37.39) | .03a |
| cirAEseverity | 1.69 (90.67-4.26) | .26 |
| Arthritis | ||
| cirAE morphologic characteristic (mucositis)d | 7.96 (1.18-53.62) | .03e |
| cirAEseverity | 0.31 (0.40-2.46) | .27 |
Abbreviations: cirAE, cutaneous immune-related adverse event; GI, gastrointestinal; irAE, noncutaneous immune-related adverse event; NS, not significant; OR, odds ratio.
P values remained statistically significant with 2-sided P value beneath the critical value, as defined using the false discovery rate control method, with a false discovery rate of 0.15.14 ICI regimen was significant only in analyses of gastrointestinal irAEs and gastroenterocolitis.
Provided OR is for mucositis or psoriasiform first cirAE relative to reference category of maculopapular reaction in multivariate model adjusted for age, sex, ICI regimen (statistically significant to P < .05), and cancer type (statistically significant to P < .05).
Associations between specific morphologic characteristics of first cirAE and hepatobiliary irAE, pulmonary irAE, ocular irAE, and other irAE were not significant.
Provided OR is for psoriasiform or mucositis first cirAE relative to the reference category of maculopapular reaction in multivariate model adjusted for age, sex, ICI regimen (statistically significant to P < .05), and cancer type (statistically significant to P < .05).
The association between mucositis and rheumatologic irAE, and mucositis and arthritis had P values below the significance threshold of P < .05, but their significance did not remain after accounting for the issue of multiple hypothesis testing using the false discovery rate control method.4
Patients with psoriasis as their first cirAE trended toward increased rates of thyroiditis, but this did not reach statistical significance.
Discussion
In this retrospective cohort study, we characterized the patterns and associations between cirAEs and irAEs among a cohort of patients with cirAEs. We found that patients with cirAEs were at high risk for multimorbid toxic effects, with irAEs manifesting in nearly half of the patients. We also observed several significant associations between specific cirAE morphologic factors and irAE types. Mucositis was associated with overall irAE risk, gastrointestinal irAEs, and gastroenterocolitis. In addition, psoriasis was associated with a risk of endocrine irAEs overall. Collectively, these findings underscore an apparently elevated risk of multisystem toxic effects in patients with cirAEs and opportunities for dermatologists to enhance the screening and care of patients with potential noncutaneous irAEs.
The irAE diagnoses associated with cirAEs are some of the more significant ICI-associated toxic effects. Gastroenterocolitis necessitates suspension of ICI therapy in more than 75% of patients and is one of the most commonly fatal irAEs across regimens.15 Similarly, an ICI-mediated endocrine toxic effect commonly causes permanent organ damage necessitating ongoing hormonal supplementation.15 Because the underlying factors indicating emergence of irAEs remain poorly understood, our results suggest that patients with specific early forms of cirAEs may be more likely to develop certain forms of extracutaneous toxic effects. Although this subset of patients merits further characterization in future studies, our findings preliminarily identify a high-risk group that might benefit from earlier surveillance and referral. Close collaboration between dermatologists and oncologists in the care of these patients could support earlier identification of gastrointestinal, rheumatologic, and endocrine irAEs, potentially reducing ICI disruption and treatment-associated morbidity.4,15 These efforts may be particularly relevant for patients who develop mucositis, as the median time to onset relative to development of any irAE would be sufficient for potential intervention.
In addition, it is noteworthy that only cirAE morphologic characteristics associated with specific histopathologic patterns (ie, psoriasiform reactions and mucositis) were associated with emergence of noncutaneous irAEs.1,2,3 We did not observe any associations between the most common cirAE category in our data set (maculopapular reaction) and any irAEs by organ class or specific toxic effect. One possible explanation for these findings, supported by earlier work, is that the maculopapular descriptor, which is frequently reported by clinicians without dermatologic expertise, may tend to conflate certain disparate cirAE types with differing associations with extracutaneous irAEs.3 Fewer than 33% of the patients in this study were evaluated by dermatology services for their first cirAE, but more than 80% of psoriasiform diagnoses were delivered by dermatology services. Collectively, these findings suggest more frequent referral might enhance categorization of certain cirAEs, potentially supporting further elucidation of the interplay between these events and noncutaneous toxic effects.
Limitations
Limitations of the study include the small number of patients within certain cirAE and irAE subgroups and the low rate of cirAEs overall, which lessen the generalizability and power of our analyses. In addition, given known issues with underdiagnosis of both cirAEs and irAEs, our findings may skew toward the more severe subset of reactions likeliest to warrant clinical attention.2,15 In addition, although our use of electronic medical record data allowed us to capture cirAEs across care settings rather than solely the small number of events referred to dermatology services, this approach may have limited the granularity of cirAE characterization, introduced interviewer bias, and lowered approximated cirAE rates. These factors may be particularly relevant for cirAEs without visible manifestations, such as isolated pruritus.
Conclusions
Despite the constraints of this study, our findings provide further characterization of associations between specific subgroups of cirAEs and irAEs, providing information that may be valuable in the care of high-risk patients with multiple ICI-associated toxic effects. Dermatologists may play a crucial role in the care of patients with cirAEs with targeted screening for potential noncutaneous irAEs.
eTable 1. Features of All cirAE Events
eTable 2. Billing Codes and Associated Morphologies
References
- 1.Coleman E, Ko C, Dai F, Tomayko MM, Kluger H, Leventhal JS. Inflammatory eruptions associated with immune checkpoint inhibitor therapy: a single-institution retrospective analysis with stratification of reactions by toxicity and implications for management. J Am Acad Dermatol. 2019;80(4):990-997. doi: 10.1016/j.jaad.2018.10.062 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Geisler AN, Phillips GS, Barrios DM, et al. Immune checkpoint inhibitor–related dermatologic adverse events. J Am Acad Dermatol. 2020;83(5):1255-1268. doi: 10.1016/j.jaad.2020.03.132 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Sibaud V. Dermatologic reactions to immune checkpoint inhibitors. Am J Clin Dermatol. 2018;19(3):345-361. doi: 10.1007/s40257-017-0336-3 [DOI] [PubMed] [Google Scholar]
- 4.Das S, Johnson DB. Immune-related adverse events and anti-tumor efficacy of immune checkpoint inhibitors. J Immunother Cancer. 2019;7(1):306. doi: 10.1186/s40425-019-0805-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Haratani K, Hayashi H, Chiba Y, et al. Association of immune-related adverse events with nivolumab efficacy in non–small-cell lung cancer. JAMA Oncol. 2018;4(3):374-378. doi: 10.1001/jamaoncol.2017.2925 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Quach HT, Dewan AK, Davis EJ, et al. Association of anti-programmed cell death 1 cutaneous toxic effects with outcomes in patients with advanced melanoma. JAMA Oncol. 2019;5(6):906-908. doi: 10.1001/jamaoncol.2019.0046 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Min Lee CK, Li S, Tran DC, et al. Characterization of dermatitis after PD-1/PD-L1 inhibitor therapy and association with multiple oncologic outcomes: a retrospective case-control study. J Am Acad Dermatol. 2018;79(6):1047-1052. doi: 10.1016/j.jaad.2018.05.035 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Sanlorenzo M, Vujic I, Daud A, et al. Pembrolizumab cutaneous adverse events and their association with disease progression. JAMA Dermatol. 2015;151(11):1206-1212. doi: 10.1001/jamadermatol.2015.1916 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Phillips GS, Wu J, Hellmann MD, et al. Treatment outcomes of immune-related cutaneous adverse events. J Clin Oncol. 2019;37(30):2746-2758. doi: 10.1200/JCO.18.02141 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Molina GE, Allen IM, Hughes MS, et al. Prognostic implications of co-occurring dermatologic and gastrointestinal toxicity from immune checkpoint inhibition therapy for advanced malignancies: a retrospective cohort study. J Am Acad Dermatol. 2020;82(3):743-746. doi: 10.1016/j.jaad.2019.07.049 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Freeman-Keller M, Kim Y, Cronin H, Richards A, Gibney G, Weber JS. Nivolumab in resected and unresectable metastatic melanoma: characteristics of immune-related adverse events and association with outcomes. Clin Cancer Res. 2016;22(4):886-894. doi: 10.1158/1078-0432.CCR-15-1136 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Hua C, Boussemart L, Mateus C, et al. Association of vitiligo with tumor response in patients with metastatic melanoma treated with pembrolizumab. JAMA Dermatol. 2016;152(1):45-51. doi: 10.1001/jamadermatol.2015.2707 [DOI] [PubMed] [Google Scholar]
- 13.Day D, Hansen AR. Immune-related adverse events associated with immune checkpoint inhibitors. BioDrugs. 2016;30(6):571-584. doi: 10.1007/s40259-016-0204-3 [DOI] [PubMed] [Google Scholar]
- 14.Glickman ME, Rao SR, Schultz MR. False discovery rate control is a recommended alternative to Bonferroni-type adjustments in health studies. J Clin Epidemiol. 2014;67(8):850-857. doi: 10.1016/j.jclinepi.2014.03.012 [DOI] [PubMed] [Google Scholar]
- 15.Thompson JA, Schneider BJ, Brahmer J, et al. Management of immunotherapy-related toxicities, version 1.2019. J Natl Compr Canc Netw. 2019;17(3):255-289. doi: 10.6004/jnccn.2019.0013 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable 1. Features of All cirAE Events
eTable 2. Billing Codes and Associated Morphologies