Skip to content
All organs

CHECKPOINT-INHIBITOR TOXICITY

Skin immune-related adverse events

The commonest and usually the earliest irAEs — mostly inflammatory eruptions, pruritus and vitiligo, rarely a reason to stop the drug. Blisters, mucosal involvement or bullous formation are the finding that changes the pathway.

The only corticosteroid this organ renders is topical, and it is scoped to pruritus without rash. No chapter cited here states a systemic dose or taper for skin toxicity, so every systemic rung says the dose is not carried — that is a gap in the atlas, not advice that steroids are unnecessary. Educational reference, not medical advice.

Anchored on the ASCO, ESMO and SITC chapters for this organ. Every source is named beside the position it supports, and labelled with what kind of source it is.

At a glance

ICI-associated inflammatory and bullous dermatoses, pruritus and vitiligo

Also called: cutaneous irAE · cirAE · ICI rash · maculopapular eruption · morbilliform eruption · eczematous dermatitis · lichenoid dermatitis · bullous pemphigoid · pemphigus · pruritus · vitiligo · checkpoint inhibitor rash

The most commonly reported irAEs and often the earliest, usually topical and rarely a reason to stop the drug — but the eruption has to be looked at rather than counted: blisters, mucosal involvement or bullous formation mean a biopsy, an antibody panel and an ICI interruption before anything else, because the answer decides between a bullous dermatosis treated here and a severe cutaneous adverse reaction treated on the other card.

commonroutineonset: weeks 1–6

Reported frequency

  • up to 71.5%any-grade cutaneous toxicity across ICPi therapy

    of individuals across ICPi therapy · ASCO cutaneous chapter. The figure is for cutaneous toxicity as a whole; rash, pruritus with and without eruption, and vitiligo are the phenotypes ASCO names as the most common, introduced by 'including' rather than as the whole of what is counted. It is an upper bound, not a central estimate

    guideline · 34724392

  • >50% for all gradesany-grade cutaneous adverse events

    of patients on ICI therapy · ESMO skin chapter, which pairs the figure with its own qualifier that these events are rarely severe and usually do not impair treatment continuation. Carried beside ASCO's upper bound rather than averaged with it — they are two guidelines' summary figures over different literatures

    guideline · 36270461

  • 16.7%any-grade rash

    of patients in ICI-containing regimens pooled across 174 published trials · Melanoma, renal cell carcinoma, non-small cell lung cancer and urothelial carcinoma; ICI monotherapy and combinations with chemotherapy, antiangiogenic agents or another ICI

    pooled meta-analysis · 38254829

  • 18.0%any-grade pruritus

    of patients in ICI-containing regimens pooled across 174 published trials · Same pooled trial set as the rash figure above; the two are counted as separate events

    pooled meta-analysis · 38254829

  • 15% (0% grade ≥3) with anti-PD-(L)1, 25% (1% grade ≥3) with anti-CTLA-4, and 34% (2% grade ≥3) with combination ICIsany-grade pruritus by ICI class, with the grade 3 or higher share in brackets

    of patients treated with ICIs, by class, as reported in SITC's skin chapter · SITC states these in the order anti-PD-(L)1, anti-CTLA-4, combination — the reverse of the first two classes in its rash sentence, which is why each set is transcribed with its own order rather than mapped onto a shared one. SITC also notes pruritus may present with no obvious skin lesions in 50% of cases. Filed under any-ici because the row spans three strata

    guideline · 34172516

  • 6.6%any-grade vitiligo

    of patients in ICI-containing regimens pooled across 174 published trials · Same pooled trial set. Carried separately because vitiligo is a different phenotype with a different course from an inflammatory eruption

    pooled meta-analysis · 38254829

  • 4% with ipilimumab, 9% with nivolumab and 6% with pembrolizumabvitiligo by agent

    of patients treated with each agent in a systematic review cited by SITC · SITC reports this per agent rather than per class; ipilimumab is anti-CTLA-4 and the other two anti-PD-1, so the row spans strata and is filed under any-ici. ASCO adds that vitiligo is reported primarily in patients with melanoma and can manifest months after treatment starts

    guideline · 34172516

  • an estimated 1%bullous pemphigoid

    of patients receiving anti-PD-(L)1 therapy · SITC's estimate. It is the only frequency this card carries for its most consequential non-SCAR phenotype, and the one whose confirmation opens the rituximab recommendation. Filed under pd1 because the vocabulary has no combined PD-1/PD-L1 member; SITC's scope is both

    guideline · 34172516

  • 25.1%any cutaneous irAE, real-world claims cohort

    of 8637 ICI-treated patients · United States national insurance claims database, retrospective, with 8637 matched controls. A claims denominator rather than a trial one, and the study notes it had no access to patient chart data

    cohort · 33819538

  • 52.3%any cutaneous adverse event on combination immunotherapy

    of 155 patients with metastatic melanoma consecutively treated with combination checkpoint inhibitors · Single-institution consecutive cohort treated between 2012 and 2017 with ipilimumab plus nivolumab or pembrolizumab for stage IV or unresectable stage III melanoma. Eighty-one patients experienced a total of 92 events

    cohort · 34724197

Severity and class

  • 3% or fewersevere cutaneous toxicity, grade 3 or higher per CTCAE

    of individuals receiving ICPi monotherapy · ASCO cutaneous chapter, monotherapy only. SITC's rash rates on this card put the grade-3-or-higher share at 5% for combination therapy — a different population, not a competing estimate of the same one, so no single figure is offered for both. ESMO separately describes maculopapular rash as severe (grade >3) in fewer than 5% of cases

    guideline · 34724392, 34172516

  • 23% (1% grade ≥3) with anti-CTLA-4, 10% (<1% grade ≥3) with anti-PD-(L)1, and 41% (5% grade ≥3) with combination ICI therapyany-grade rash by ICI class, with the grade 3 or higher share in brackets

    of patients treated with ICIs, by class, as reported in SITC's skin chapter · SITC states these three rates in one sentence in the order anti-CTLA-4, anti-PD-(L)1, combination. Its pruritus sentence uses a DIFFERENT class order, so the two sets must never be read off one order; the pruritus rates are carried in their own row below. ASCO reports the same direction independently — cutaneous irAEs more frequently observed with CTLA-4 inhibitors alone or in combination with anti-PD-1 agents — and a meta-analysis of 174 trials adds that the excess tracks higher doses of ipilimumab

    guideline · 34172516, 34724392, 38254829

Onset

ASCO states a median time to onset of skin toxicity of 4 weeks with a broad range of 2 to 150 weeks, and puts cutaneous toxicity among the earliest irAEs — typically within 3 weeks of starting ipilimumab and within 6 weeks of starting anti-PD-1 — while adding that onset can be delayed, even after therapy is complete. SITC gives an average of 3-4 weeks for dermatological toxicity overall and 2-5 weeks for rash specifically; ESMO puts non-specific maculopapular rashes in the first 6 weeks. The timing is phenotype-dependent: ASCO describes inflammatory dermatoses presenting within the first one to two cycles and vitiligo manifesting months after initiation, which is the likely reason the real-world claims cohort, counting every phenotype, reports a later median of 113 days. The combination-therapy melanoma cohort reports a median 21 days with a range of 0-341 days. Each figure is carried in the units its source used.

Presentation

  • · Rash, pruritus with or without an eruption, and vitiligo — the three ASCO names as the most common cutaneous toxicities
  • · Inflammatory dermatitis covers erythema multiforme, lichenoid, eczematous, psoriasiform, morbilliform and palmoplantar erythrodysesthesia; itch with or without rash, macules, papules or plaques, and loss of skin pigmentation are ASCO's presenting symptoms
  • · Bullous dermatoses present with bullae, persistent urticaria or erosions on the skin or mucosal surfaces — ASCO's second category, and the one that most often needs a biopsy to name
  • · ESMO: non-specific maculopapular rash is the commonest morphology, usually involves under 30% of body surface area, and is considered severe (grade >3) in fewer than 5% of cases; pruritus may be the sole manifestation of a bullous pemphigoid
  • · Eczematous dermatitis, morbilliform eruption, vitiligo and pruritus without rash were the phenotypes recorded in a combination-therapy melanoma cohort
  • · Pruritus, mucositis, erythroderma, maculopapular eruption, vitiligo, lichen planus, bullous pemphigoid, Grover disease and drug eruption were all significantly more frequent in ICI recipients than in matched controls
  • · Mostly reversible: median time to resolution 50 days, range 1-352 days, in the combination cohort
  • · Rarely a reason to stop — in that cohort 4 patients (2.6%) discontinued because of a cutaneous adverse event, against 49 (31.6%) who stopped for other immune-related adverse events

Differential

  • · A severe cutaneous adverse reaction. Blisters, mucosal involvement or bullous formation are SITC's trigger to suspect lichenoid rash, pemphigus, pemphigoid or SJS/TEN, to biopsy and to interrupt the ICI — all of which happen on THIS card. The biopsy and antibody panel are what separate them: pemphigus, bullous pemphigoid and lichenoid dermatitis are managed here, where their steroid-sparing agents are indexed; only confirmed SJS/TEN, DRESS or AGEP moves to the severe-reaction card
  • · Progression rather than a fixed diagnosis — SITC describes SJS/TEN as sometimes appearing progressively out of a less severe skin toxicity, such as lichenoid eruptions that fail to respond to typical therapies, so a failing lichenoid rash is re-examined and not just re-treated
  • · Another drug: ESMO asks that the relationship between the ICI and the skin event be evaluated and confirmed where possible, since these patients are usually on several medications
  • · A pre-existing dermatosis flaring — ESMO asks specifically about a history of psoriasis or an autoimmune disease with a skin manifestation
  • · Infection: SITC treats a rash that does not respond to immunosuppression as a reason to suspect one and to culture

Work-up

  • Full examination of the skin and mucosae before ICI therapy starts · All patients

    Baseline documentation is what makes a later lesion interpretable, and ESMO puts it before the first dose rather than at the first rash.

    36270461

  • History of psoriasis or autoimmune disease with a skin manifestation, and review of concomitant medications · All patients

    Establishes attribution: whether this is the ICI, another drug, or a pre-existing dermatosis flaring.

    36270461

  • CBC with differential, CMP, percentage body surface area involved, assessment for bullous formation or mucosal involvement, and a history of allergy or atopy · All patients

    SITC's workup for a potential ICI-related rash below grade 3. The special-features question is the one that changes the pathway, and it is asked of every rash rather than only of severe ones.

    34172516

  • Severity assessment with an explicit decision on specialist advice or referral · All patients

    ESMO makes the referral question part of the assessment, and states its purpose plainly: repeated severity evaluation exists to exclude the rare severe entities.

    36270461

  • Dermatology consultation · Grade 2+

    SITC recommends it when a rash does not respond to topical or oral corticosteroids, is grade 3 or higher, or is intolerable — and again for a grade 2 or higher event that recurs after a steroid taper, where it is paired with the steroid-sparing agents rather than offered instead of them.

    34172516

  • Skin biopsy with direct immunofluorescence, plus serum desmoglein 1 and 3 and bullous pemphigoid antigen 1 and 2 antibodies and indirect immunofluorescence on salt-split skin · If atypical

    The test set SITC attaches to blisters, mucosal involvement or bullous formation, and the step that decides treatment: pemphigus, bullous pemphigoid and lichenoid dermatitis are managed on this card with the matching steroid-sparing agent, while SJS/TEN, DRESS or AGEP moves to the severe-reaction card. ESMO adds what confirms a bullous pemphigoid — C3 and IgG deposits on the basal membrane by direct immunofluorescence, with anti-basal membrane antibodies on serology.

    34172516

  • Cultures of the rash · If refractory

    SITC's rule for a rash that does not respond to immunosuppressive therapy: consider infection.

    34172516

Grade ladder

Grade 1

Continue the ICIOutpatient

Grade 1 — limited rash or pruritus. The body-surface-area thresholds CTCAE uses are not reproduced: no chapter quoted for this entry states them, and a threshold is not a figure to supply from memory. The care setting on this rung is the atlas's reading, not a guideline's — no chapter quoted here states an admission threshold below grade 4.

TopicalHigh-potency topical corticosteroids and GABA agonists — SITC's option for pruritus without rash

SITC states this for pruritus WITHOUT rash (LE: 4) and the scope is kept: it is not generalised into a topical plan for every grade 1 eruption. The potency word belongs to that indication — for an inflammatory rash, no chapter quoted here states a potency class, schedule or duration. Separately, and descriptively rather than as guidance, most cutaneous adverse events in the combination-immunotherapy cohort resolved after patients entered the maintenance phase and were treated with oral antihistamines with or without topical steroids; that cohort does not index treatment by CTCAE grade.

  • · Examine skin and mucosae rather than triaging from a description
  • · Assess the body surface area involved and look specifically for bullous formation or mucosal involvement — SITC asks this of every rash below grade 3
  • · Confirm the ICI is the likely cause — these patients are usually on several medications
  • · Ask about psoriasis or another skin-manifesting autoimmune disease

Sources differ here

ICI action

  • ASCO 2021 (society guideline)While management varies according to the organ system affected, in general, ICPi therapy should be continued with close monitoring for grade 1 toxicities
  • SITC 2021 (society guideline)Rash with blisters, mucosal involvement, or bullous formation should trigger suspicion of lichenoid rash, pemphigus, pemphigoid, or SJS/TEN and follow-up testing ... ICI therapy should be interrupted until resolution or significant improvement of symptoms — a morphological trigger, not a grade one, so it applies to a limited eruption

Escalate when: Blisters, mucosal involvement or bullous formation. That triggers the biopsy, direct immunofluorescence and serum antibody panel, and an ICI interruption until resolution or significant improvement — on this card, whatever the grade. Where the biopsy returns pemphigus, bullous pemphigoid or a lichenoid dermatosis the patient stays here and the steroid-sparing agent follows the diagnosis; only confirmed SJS/TEN, DRESS or AGEP moves to the severe cutaneous adverse reaction card. Mucosal involvement at ocular, oral or anogenital sites, skin pain or sloughing, or a lichenoid eruption failing usual therapy, are SITC's markers that the reaction may be heading that way.

Grade 2

Hold the ICIOutpatient

Grade 2 — the level at which SITC's general framework withholds the ICI and at which its steroid-sparing recommendation applies to rash or pruritus recurring after a steroid taper. No body-surface-area threshold is carried, and the care setting is the atlas's reading rather than a stated threshold.

Dose not carried

No chapter quoted for this entry states a systemic corticosteroid dose, route, taper window or taper-start trigger for skin toxicity. SITC's framework withholds the ICI and treats grade 2 events with corticosteroids 'depending on the toxicity in question', and its skin chapter names recurrence after a steroid taper, so systemic steroids are clearly contemplated — but no number is attached to them, and ASCO's cross-organ figures are not a skin-chapter source (D-STEROID). The topical option is on the grade 1 rung; the steroid-sparing agents below are what the chapters do specify at this grade.

Second line

  • RituximabSITC recommends a dermatological consultation or a steroid-sparing agent for grade 2 or higher dermatological events (rash, pruritus) that recur after a steroid taper, and names rituximab for pemphigus (LE: 1) or bullous pemphigoid (LE: 4). Indexed to the biopsy-confirmed dermatosis, not to the grade — which is why the biopsy, direct immunofluorescence and antibody panel come first.
  • DupilumabSITC's steroid-sparing option for eczema (LE: 1), for a grade 2 or higher event recurring after a steroid taper.
  • InfliximabSITC's steroid-sparing option for lichenoid rash, for a grade 2 or higher event recurring after a steroid taper. A lichenoid eruption that fails usual therapy is also SITC's described route by which SJS/TEN can appear progressively, so failure here is a re-examination, not only a next drug.
  • OmalizumabSITC's steroid-sparing option for urticaria or pruritus (LE: 1), for a grade 2 or higher event recurring after a steroid taper.
  • · Dermatology consultation — SITC pairs it with the steroid-sparing agents rather than offering it as an alternative to them
  • · Establish the dermatosis before choosing the agent: the four options above are indexed to different diagnoses, which is what the biopsy and antibody panel are for
  • · ASCO's general rule at this grade is suspension with consideration of resuming once symptoms revert to grade 1 or lower

Escalate when: Recurrence after a steroid taper, which is SITC's trigger for dermatology consultation or a steroid-sparing agent; a rash that does not respond to topical or oral corticosteroids or is intolerable, which is its consultation trigger; or new blisters, mucosal involvement or bullous formation, which sends the patient to biopsy.

Grade 3

Hold the ICIConsider admission

Grade 3 — from ASCO's cross-toxicity ladder rather than a cutaneous definition: grade 3 toxicities generally warrant suspension of ICPis and the initiation of high-dose corticosteroids. ASCO reports grade 3 or higher cutaneous toxicity in 3% or fewer individuals on monotherapy. SITC adds that a grade 3 or higher rash is itself a reason for dermatological consultation and possible skin biopsy. The care setting is the atlas's reading, not a stated threshold.

Dose not carried

ASCO calls for high-dose corticosteroids at grade 3 across organ systems and names no dose in that sentence; no cutaneous chapter quoted here supplies one, and a cross-organ figure is not a skin-chapter source (D-STEROID). The rung states that the dose is not carried — not that no steroid is indicated.

Second line

  • RituximabSITC recommends a dermatological consultation or a steroid-sparing agent for grade 2 or higher dermatological events (rash, pruritus) that recur after a steroid taper, and names rituximab for pemphigus (LE: 1) or bullous pemphigoid (LE: 4). Indexed to the biopsy-confirmed dermatosis, not to the grade — which is why the biopsy, direct immunofluorescence and antibody panel come first.
  • DupilumabSITC's steroid-sparing option for eczema (LE: 1), for a grade 2 or higher event recurring after a steroid taper.
  • InfliximabSITC's steroid-sparing option for lichenoid rash, for a grade 2 or higher event recurring after a steroid taper. A lichenoid eruption that fails usual therapy is also SITC's described route by which SJS/TEN can appear progressively, so failure here is a re-examination, not only a next drug.
  • OmalizumabSITC's steroid-sparing option for urticaria or pruritus (LE: 1), for a grade 2 or higher event recurring after a steroid taper.
  • · Dermatology consultation and possible skin biopsy — SITC's stated workup at this grade
  • · Re-examine mucosae: involvement there is a red flag about diagnosis, not a severity gradient
  • · If the rash is not responding to immunosuppression, culture it: SITC treats non-response as a reason to suspect infection

Escalate when: Failure to respond to immunosuppressive therapy, which SITC reads as possible infection, or the appearance of blisters, mucosal involvement or bullous formation.

Grade 4

Discontinue permanentlyAdmit

Grade 4 — ASCO's cross-toxicity rule: permanent discontinuation of ICPis is recommended with grade 4 toxicities, except for endocrinopathies controlled by hormone replacement, an exception that does not apply here. This is also the one rung whose care setting is sourced: ASCO states that grade 4 toxicities will likely require hospitalization for management of infection risk, symptoms, wound care and potential supplemental nutrition.

Dose not carried

No cutaneous chapter quoted here states a grade 4 corticosteroid dose, route or taper. ASCO's grade 4 statement is about discontinuing the ICPi, not about dosing steroids.

Second line

  • RituximabSITC recommends a dermatological consultation or a steroid-sparing agent for grade 2 or higher dermatological events (rash, pruritus) that recur after a steroid taper, and names rituximab for pemphigus (LE: 1) or bullous pemphigoid (LE: 4). Indexed to the biopsy-confirmed dermatosis, not to the grade — which is why the biopsy, direct immunofluorescence and antibody panel come first.
  • DupilumabSITC's steroid-sparing option for eczema (LE: 1), for a grade 2 or higher event recurring after a steroid taper.
  • InfliximabSITC's steroid-sparing option for lichenoid rash, for a grade 2 or higher event recurring after a steroid taper. A lichenoid eruption that fails usual therapy is also SITC's described route by which SJS/TEN can appear progressively, so failure here is a re-examination, not only a next drug.
  • OmalizumabSITC's steroid-sparing option for urticaria or pruritus (LE: 1), for a grade 2 or higher event recurring after a steroid taper.
  • · Hospitalisation for infection risk, symptom control, wound care and possible supplemental nutrition — ASCO's grade 4 statement for cutaneous toxicity
  • · Re-read the eruption against SITC's red flags: a grade 4 skin event is where a severe cutaneous adverse reaction is most likely to have been missed
  • · Culture before assuming steroid failure is immunological

Escalate when: Already the top rung of this card. Confirmed SJS/TEN, DRESS or AGEP redirects to the severe cutaneous adverse reaction card, and non-response still raises infection.

Rechallenge

reasonable

SITC states it directly: patients who have experienced a grade 3 ICI-related rash may be re-challenged with ICIs, and its general condition is that signs and symptoms have resolved or are controlled on 10 mg or less of prednisone equivalent per day. ESMO's chapter says cutaneous events are rarely severe and usually do not impair treatment continuation, and in the combination cohort 4 patients (2.6%) discontinued because of a cutaneous adverse event against 49 (31.6%) who stopped for other irAEs. Note that this is a genuinely more permissive tier than the shared rechallenge calculator returns: evaluateRechallenge() treats a grade 3-4 event as a hard stop, which is derived from ICI-AKI, where the organ can be lost. SITC's skin sentence is what licenses the difference here, and it does not extend to a severe cutaneous adverse reaction — that is a different card with a different answer.

  • · Signs and symptoms resolved, or controlled with 10 mg or less of prednisone or equivalent per day — SITC's general condition for re-challenge
  • · Symptoms resolved or significantly improved, where a red-flag rash interrupted therapy — SITC's condition for restarting after that interruption
  • · Symptoms at grade 1 or lower, which is where ASCO's general ladder considers resumption
  • · Re-examine skin and mucosae, and confirm the eruption was never one of the red-flag morphologies — or, if it was, that the biopsy returned a dermatosis rather than a severe cutaneous adverse reaction

Deliberately not carried

  • guidelinePmidsLeft empty although all three anchors are rows in guidelines.ts. Their stored keyRecommendation strings are renal-scoped, so linking them here would surface kidney dosing advice on a skin card — the substitution 19.5 exists to block. The cutaneous chapters are carried in citations with their own quotes instead.
  • ladder[].steroid dose and taperNo anchor chapter quoted for this organ states a systemic corticosteroid dose, route, taper window, taper-start trigger or step cadence for skin toxicity. ASCO's grade 3 wording is 'high-dose corticosteroids' with no number attached, and its numbered cross-organ figures are not a cutaneous statement (D-STEROID). Grades 2 to 4 therefore carry kind: unsourced, there is no systemic rung and so no taper object at all. This is the gap a dermatology reviewer should look at first.
  • ladder[].hospitalization, grades 1-3ASCO states a care setting for grade 4 only ('will likely require hospitalization'), which is what the grade 4 rung carries. Nothing quoted here states an admission threshold at grades 1-3; the schema requires the field, so those three values are the conservative reading of the grade ladder and each rung's gradeDefinition says so, rather than the caveat living only in this list.
  • incidence[].agentClass for the multi-stratum rowsThe rows carrying two or three class-specific rates in one value string are filed under any-ici, because agentClass takes one value and no member describes 'several strata at once'. The rates themselves name their classes in the value, so nothing is lost — but the row cannot be filtered by class. Single-stratum anti-PD-(L)1 rates are filed under pd1-pdl1 and are not affected.
  • onset.bucketThe bucket carries ASCO's median of 4 weeks, which is where most of these events sit and what the client index can represent. It does not represent the tail: the same sentence gives a range out to 150 weeks, vitiligo can appear months in, and the real-world claims cohort's median is 113 days. A late eruption is not evidence against attribution.
  • onset, per phenotypeASCO describes the gradient — inflammatory dermatoses within the first one to two cycles, vitiligo months after initiation — and SITC gives rash 2-5 weeks, but no source quoted here states a median per phenotype. The structured fields therefore carry the all-cutaneous median and range, with the gradient in the note.
  • incidence, cutaneous adverse event subtypes in the combination cohortThe cohort lists eczematous dermatitis, morbilliform eruption, vitiligo and pruritus without rash with percentages, but does not state whether the denominator is the 155 patients or the 92 events. The phenotypes are named in `presentation`; the percentages are not carried, because picking a denominator would be the atlas's arithmetic rather than the paper's.
  • mortalityNo source cited here reports a case-fatality rate for inflammatory or bullous cutaneous irAEs, and none is derived. The discontinuation figures are a tolerability measure, not a mortality one, and are not repurposed as one.
  • rechallenge.recurrenceNo source cited here reports a recurrence rate for cutaneous irAEs after resuming an ICI. The colitis and hepatitis cards carry recurrence fractions; this one has none, and an absent field is the honest rendering rather than an em-dash implying zero.
  • prognostic association with tumour response and survivalThe meta-analysis correlates cirAE incidence with response rate and overall survival, and the combination cohort reports a longer time to next treatment in patients with a cutaneous event. Both are study-level or association findings; rendering them on a clinical card invites reading them as an individual prognosis, so they are not carried.
  • gradeDefinition thresholdsASCO's per-organ grade tables are images that did not survive extraction from the supplied chapter archives, and no body-surface-area threshold appears in the narrative text quoted here — ESMO's 'usually involve <30% of body surface area' describes the typical maculopapular rash, not a CTCAE band. The rungs describe what each guideline does at that grade instead of restating CTCAE.

Sources

  • Schneider BJ, et al. (2021) Management of Immune-Related Adverse Events in Patients Treated With Immune Checkpoint Inhibitor Therapy: ASCO Guideline UpdateCited for: The three-way classification of cutaneous irAEs that this card's scope rests on, the median and range of onset for skin toxicity, what an inflammatory dermatitis and a bullous dermatosis look like, the any-grade cutaneous frequency, the class direction, the onset window by agent, the grade-3-or-higher share on monotherapy, the phenotype-dependent timing, the grade 4 hospitalization statement, and the cross-toxicity grade ladder. The FIRST block is ASCO's cross-organ Recommendations summary — it also appears verbatim in the abstract, and it is cited here as ASCO's general default, not as a cutaneous recommendation. Everything after it is the cutaneous chapter.Supporting text: the full text + the PubMed abstract (checkable at the link above)
  • Haanen J, et al. (2022) Management of toxicities from immunotherapy: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-upCited for: The all-grade cutaneous frequency and the statement that these events are rarely severe, the commonest morphology and its 6-week window, pruritus as the sole manifestation of a bullous pemphigoid, the body-surface-area and severity description of maculopapular rash, the immunofluorescence and serology that confirm bullous pemphigoid, the repeated severity evaluation whose purpose is to exclude the severe entities, and the four pre-treatment and attribution recommendations. Bullet order is the source's.Supporting text: the full text
  • Brahmer JR, et al. (2021) Society for Immunotherapy of Cancer (SITC) clinical practice guideline on immune checkpoint inhibitor-related adverse eventsCited for: The general grade 1 and grade 2 ICI actions and the general rechallenge condition (SITC's cross-organ framework, which its skin chapter explicitly writes within); the average onset of dermatological toxicity; the class-stratified rash, pruritus and vitiligo rates with their grade-3-or-higher shares; the estimated pemphigoid rate; that SJS/TEN is rare, is accompanied by mucosal involvement, and may appear progressively out of a lichenoid eruption that fails usual therapy; the treatment of pruritus without rash; the workup by grade; the red-flag morphology with its biopsy and antibody panel and the ICI interruption it carries; the consultation triggers; the steroid-sparing agents by dermatosis; the grade 3 rash rechallenge sentence; and the rule that non-response raises infection.Supporting text: the full text
  • Curkovic NB, Bai K, Ye F, Johnson DB (2024) Incidence of Cutaneous Immune-Related Adverse Events and Outcomes in Immune Checkpoint Inhibitor-Containing Regimens: A Systematic Review and Meta-Analysis.Cited for: Pooled incidence of the three commonest cutaneous irAEs as separate entities, the regimens and tumour types pooled, and the direction of the class skew including the ipilimumab dose-response.Supporting text: the PubMed abstract (checkable at the link above)
  • Wongvibulsin S, et al. (2022) Epidemiology and risk factors for the development of cutaneous toxicities in patients treated with immune-checkpoint inhibitors: A United States population-level analysis.Cited for: Real-world cumulative incidence and median onset of cutaneous irAEs against a matched control arm, the phenotypes that were significantly more frequent than in controls, the higher-risk groups, and the study's own stated limitation. Section boundaries in the structured abstract are marked the same way as any other elision.Supporting text: the PubMed abstract (checkable at the link above)
  • Patel AB, et al. (2022) Cutaneous adverse events in 155 patients with metastatic melanoma consecutively treated with anti-CTLA4 and anti-PD1 combination immunotherapy: Incidence, management, and clinical benefit.Cited for: Cohort definition, the share of patients with a cutaneous adverse event on combination therapy, median times to onset and resolution, what the events were treated with, and how rarely they stopped the immunotherapy. The cohort size of 155 is the record's title, which is stored verbatim on this entry.Supporting text: the PubMed abstract (checkable at the link above)

Severe cutaneous adverse reactions (SJS/TEN, DRESS, AGEP)

Also called: SCAR · Stevens-Johnson syndrome · SJS · toxic epidermal necrolysis · TEN · Lyell disease · DRESS · drug reaction with eosinophilia and systemic symptoms · drug-induced hypersensitivity syndrome · AGEP · acute generalized exanthematous pustulosis

ASCO's third category of cutaneous irAE, separate from the inflammatory and bullous dermatoses and managed separately. ESMO makes recognising early DRESS, Lyell disease and Stevens-Johnson syndrome a competence requirement; SITC describes SJS/TEN as sometimes rapid and sometimes emerging out of a lichenoid eruption that stops responding, which is why the previous card sends a confirmed diagnosis here rather than a morphology.

rarecriticalonset: any time on therapy

Onset

No source cited here reports an onset window specific to severe cutaneous adverse reactions. ASCO's cutaneous timing statement covers skin toxicity as a whole: often the earliest observed irAEs, typically within 3 weeks of starting ipilimumab and within 6 weeks of starting anti-PD-1, with onset that can be delayed even after therapy is complete. SITC adds a second timing pattern that is specific to this entity but not a window — SJS/TEN may be rapid and acute, or may appear progressively out of a less severe skin toxicity such as a lichenoid eruption that fails usual therapy. The bucket does not narrow either statement.

Presentation

  • · Fever, widespread rash and skin pain
  • · Skin sloughing and erosions
  • · Facial or upper-extremity oedema
  • · Pustules — acute generalized exanthematous pustulosis is grouped within SCAR by ASCO, which notes it is not always severe
  • · Mucosal involvement at ocular, oral or anogenital sites — SITC states this accompanies SJS/TEN, and it is a diagnostic feature rather than a measure of extent
  • · Two tempos, not one: SITC describes SJS/TEN as either rapid and acute, or appearing progressively out of a less severe skin toxicity such as lichenoid eruptions that fail to respond to typical therapies

Differential

  • · The bullous and lichenoid dermatoses SITC groups under the same red flag — lichenoid rash, pemphigus and pemphigoid — which the biopsy, direct immunofluorescence and serum antibody panel separate from SJS/TEN. ASCO classes them as a different category of cutaneous irAE, and they are managed on the inflammatory and bullous dermatoses card, where SITC's steroid-sparing agents are indexed to them. A confirmed diagnosis is what moves a patient between the two cards, not a morphology
  • · Infection: a rash that does not respond to immunosuppressive therapy should raise it, and SITC asks for cultures
  • · Another drug — ESMO requires the relationship between the ICI and the skin event be evaluated and confirmed where possible, since these patients are usually on several medications
  • · A pre-existing dermatosis, which ESMO asks about by history before therapy starts

Work-up

  • Full examination of the skin and mucosae before ICI therapy starts · All patients

    ESMO puts it before the first dose. Mucosal involvement is the finding that changes the diagnosis, and it can only be called new against a documented baseline.

    36270461

  • Skin biopsy with direct immunofluorescence studies · If atypical

    SITC's response to blisters, mucosal involvement or bullous formation. It is what separates SJS/TEN from pemphigus and pemphigoid, which are managed on the other card and with different drugs.

    34172516

  • Serum antibodies to desmogleins 1 and 3 and to bullous pemphigoid antigens 1 and 2 · If atypical

    The serological half of SITC's panel, distinguishing the autoimmune blistering diseases from SJS/TEN.

    34172516

  • Indirect immunofluorescence on salt-split skin · If atypical

    Completes SITC's follow-up testing set for a rash with blisters, mucosal involvement or bullous formation.

    34172516

  • Examination of ocular, oral and anogenital mucosae · All patients

    SITC names these three sites as where the mucosal involvement that accompanies SJS/TEN is found. ESMO's requirement that early Lyell disease and Stevens-Johnson syndrome be recognised is what this examination serves.

    34172516

  • Severity assessment with an explicit decision on specialist advice or referral · All patients

    ESMO states the purpose of repeated severity evaluation directly: to exclude the rare severe entities, which then need immediate interruption of the ICI, specialist treatment and monitoring.

    36270461

  • Cultures of the rash · If refractory

    SITC's rule when a rash fails to respond to immunosuppressive therapy: consider infection.

    34172516

Grade ladder

Grade 1

Hold the ICIConsider admission

Grade 1 — no chapter quoted here gives a CTCAE definition for a severe cutaneous adverse reaction, and none is supplied. What the chapters state is a diagnosis-driven rule rather than a graded one: ESMO requires immediate interruption of ICI therapy with specialist treatment and monitoring for these entities, and SITC interrupts until resolution or significant improvement once blisters, mucosal involvement or bullous formation appear. The care setting on this rung is the atlas's reading; ASCO states one at grade 4 only.

Dose not carried

No chapter quoted for this entry states a corticosteroid dose, route, taper window, taper-start trigger or step cadence for severe cutaneous adverse reactions. ASCO's cross-toxicity ladder calls for high-dose corticosteroids at grade 3 without naming a dose in that sentence, and a cross-organ figure is not a cutaneous statement (D-STEROID). Nothing is carried across.

  • · Interrupt the ICI — ESMO's word for these entities is immediate, and SITC's interruption runs until resolution or significant improvement; neither is conditioned on grade
  • · ASCO's general ladder continues an ICPi at grade 1; this card follows the entity-specific instruction instead, because the trigger here is what the eruption is rather than how much of the body it covers
  • · Biopsy with direct immunofluorescence and the serum antibody panel before treating a morphology as a diagnosis

Sources differ here

ICI action

  • ASCO 2021 (society guideline)While management varies according to the organ system affected, in general, ICPi therapy should be continued with close monitoring for grade 1 toxicities
  • SITC 2021 (society guideline)Rash with blisters, mucosal involvement, or bullous formation should trigger suspicion of lichenoid rash, pemphigus, pemphigoid, or SJS/TEN and follow-up testing ... ICI therapy should be interrupted until resolution or significant improvement of symptoms
  • ESMO 2022 (society guideline)rare severe irAEs such as Stevens–Johnson syndrome, toxic epidermal necrolysis, bullous lesions and DRESS syndrome ... necessitate immediate interruption of ICI therapy, specialist treatment and monitoring

Escalate when: Mucosal involvement at ocular, oral or anogenital sites, skin pain, sloughing or spreading blistering — or a lichenoid eruption that has stopped responding, which is SITC's described route by which SJS/TEN appears progressively. ESMO's requirement is that early DRESS, Lyell disease and Stevens-Johnson syndrome be recognised at this stage, not confirmed at the next.

Grade 2

Hold the ICIConsider admission

Grade 2 — again no SCAR-specific CTCAE definition is stated in the chapters quoted here. ASCO's cross-toxicity ladder may suspend the ICPi for most grade 2 toxicities, with consideration of resuming when symptoms revert to grade 1 or lower; the entity-specific interruption already applies. The care setting is the atlas's reading.

Dose not carried

No dose is stated for this entity in any chapter quoted here. The rung records that the dose is not carried — it does not assert that no steroid is indicated, which is what kind: none would mean.

  • · Dermatology involvement, with the diagnosis established by biopsy and serology rather than by morphology alone
  • · Keep the ICI interrupted until resolution or significant improvement
  • · If the rash does not respond to immunosuppression, culture it: SITC treats non-response as a reason to suspect infection

Escalate when: Progression of blistering, mucosal involvement or skin pain, or a rash that does not respond to immunosuppressive therapy.

Grade 3

Hold — consider permanent discontinuationAdmit

Grade 3 — no SCAR-specific severity threshold is stated in the chapters quoted here. Two ASCO statements govern the action: grade 3 toxicities generally warrant suspension of ICPis and high-dose corticosteroids, and separately, SCAR such as Stevens-Johnson syndrome and DRESS will require temporary or permanent discontinuation of therapy. The second is entity-specific, which is why this rung is 'hold, consider permanent' rather than the plain hold its sibling card carries at grade 3.

Dose not carried

ASCO calls for high-dose corticosteroids at grade 3 across organ systems and names no dose in that sentence; no cutaneous chapter quoted here supplies one for this entity. The rung states that the dose is not carried rather than importing a figure from another organ (D-STEROID).

  • · Admit, and involve dermatology
  • · Re-examine the ocular, oral and anogenital mucosae — SITC's named sites, part of the diagnosis rather than a late complication
  • · Culture before reading steroid failure as immunological

Escalate when: Progression despite immunosuppression, which by SITC also raises infection, or the appearance of grade 4 features.

Grade 4

Discontinue permanentlyAdmit

Grade 4 — ASCO's cross-toxicity rule: permanent discontinuation of ICPis is recommended with grade 4 toxicities, except for endocrinopathies controlled by hormone replacement, an exception that does not apply here. ASCO also states the care setting at this grade for cutaneous toxicity: grade 4 toxicities will likely require hospitalization for management of infection risk, symptoms, wound care and potential supplemental nutrition.

Dose not carried

No chapter quoted here states a corticosteroid dose, route or taper for grade 4 severe cutaneous adverse reactions. ASCO's grade 4 statement is about discontinuing the ICPi, not about dosing steroids.

  • · Hospitalisation for infection risk, symptom control, wound care and possible supplemental nutrition — ASCO's grade 4 statement, and the only care-setting content the chapters quoted here supply
  • · Specialist treatment and monitoring, which is what ESMO attaches to these entities alongside immediate interruption of the ICI
  • · Continue to treat non-response as a reason to look for infection, not only to escalate immunosuppression

Escalate when: Already the top rung of this card, and the point at which the atlas stops being useful: management of established SJS/TEN, DRESS or AGEP is outside what the chapters quoted here state.

Rechallenge

avoid

No source cited here states a rechallenge stance for severe cutaneous adverse reactions. SITC's sentence permitting re-challenge after a grade 3 ICI-related RASH is scoped to rash and is carried on the other card; it is not extended here. What is stated for this entity is that ESMO requires immediate interruption of ICI therapy with specialist treatment and monitoring, that SITC interrupts until resolution or significant improvement, and that ASCO expects temporary or permanent discontinuation for SCAR and permanent discontinuation at grade 4. The stance is the conservative reading of those statements rather than a sourced recommendation — which is also where the shared rechallenge calculator lands, since evaluateRechallenge() treats a grade 3-4 event as a hard stop. The atlas has no recurrence figure for this entity and no chapter here describing conditions under which rechallenge would be appropriate.

  • · Resolution or significant improvement of symptoms — SITC's condition for restarting after an interrupted skin toxicity, and the floor rather than the whole decision here
  • · A diagnosis established by biopsy, direct immunofluorescence and the serum antibody panel, not by morphology
  • · Dermatology involvement in the decision, which ESMO frames as assessing the need for specialist advice or referral

Deliberately not carried

  • guidelinePmidsLeft empty although all three anchors are rows in guidelines.ts: their stored keyRecommendation strings are renal-scoped, so linking them would surface kidney dosing advice on this card. The cutaneous chapters are carried in citations with their own quotes.
  • incidenceNo source cited here reports a rate for ICI-associated SJS/TEN, DRESS or AGEP. The list is empty rather than approximated. ASCO's headline cutaneous frequency is an ALL-CUTANEOUS figure — it counts every skin toxicity and says nothing specific about these entities — so it stays on the inflammatory and bullous dermatoses card with the sentence that states it; the same goes for SITC's estimated bullous pemphigoid rate, which belongs to a different diagnosis. Neither sentence is stored on this card's citations, and neither figure is restated here, so neither can render on a critical-tier card as though it described this entity.
  • gradeThreePlusASCO's grade-3-or-higher cutaneous figure covers all skin phenotypes on monotherapy, not severe cutaneous adverse reactions. It is carried on the other card with the sentence that states it; reusing it here would present a mixed-phenotype rate as this entity's rate, so no figure is carried.
  • mortalityNo source cited here reports a case-fatality rate for ICI-associated SJS/TEN, DRESS or AGEP, and the general SCAR literature outside the three anchors was not drawn on. The absence is a gap in this atlas, not a statement about how these reactions behave.
  • onsetNo entity-specific onset WINDOW is stated in any chapter quoted here; SITC's two tempos are a pattern, not a window. The bucket is 'any-time' because ASCO's cutaneous timing statement covers all skin toxicity and explicitly allows delayed onset, and narrowing it for this card would be inference.
  • agentClassSkewSITC reports SJS/TEN with both anti-PD-1 and anti-CTLA-4 inhibitors and quantifies neither. The class skew measured for cutaneous irAEs is for the common inflammatory phenotypes, so none is entered here.
  • ladder[].steroid dose and taperNo anchor chapter quoted here states a corticosteroid dose, route, taper window, taper-start trigger or step cadence for this entity. All four rungs carry kind: unsourced, there is no systemic rung and therefore no taper object, and neither the renal nor the hepatic figures are carried across (D-STEROID). For an entity at consequenceTier critical this is the most consequential gap on the card and is stated rather than filled.
  • secondLineSITC's steroid-sparing agents — rituximab, dupilumab, infliximab, omalizumab — are indexed to pemphigus, bullous pemphigoid, eczema, lichenoid rash and urticaria or pruritus. None of those indications is SJS/TEN, DRESS or AGEP, so neither the agents nor the sentence naming them is stored on this card; they are on the inflammatory and bullous dermatoses card, where their indications sit, and where a patient whose biopsy returns one of those diagnoses is managed.
  • supportive care for established SJS/TEN, DRESS or AGEPThe chapters quoted here state hospitalisation with wound care, infection-risk management and possible supplemental nutrition at grade 4, and specialist treatment and monitoring — and nothing about burn-unit or critical-care transfer, ocular surface management, fluid replacement or a SCORTEN-type severity score. Earlier drafts of this card asserted burn or intensive care involvement and an ICU care setting; neither is in any quote on this entry, so both are gone rather than kept as a plausible-sounding default.
  • ladder[].hospitalization, grades 1-3ASCO states a care setting for grade 4 only, which is what the grade 4 rung carries. Nothing quoted here states an admission or critical-care threshold at grades 1-3; the schema requires the field, so those values are the conservative reading for an entity ASCO defines by skin sloughing and mucosal involvement, and each rung's gradeDefinition says so.
  • crossTagsDRESS is named by ASCO as drug reaction with eosinophilia and systemic symptoms, but no chapter quoted here describes which organs are involved or how they are monitored, so no cross-organ tag is asserted from the name alone.
  • rechallenge.recurrenceNo source cited here reports what happens when an ICI is resumed after a severe cutaneous adverse reaction. The field is null and the stance says so, rather than implying the question has been studied and answered.

Sources

  • Schneider BJ, et al. (2021) Management of Immune-Related Adverse Events in Patients Treated With Immune Checkpoint Inhibitor Therapy: ASCO Guideline UpdateCited for: That SCAR is one of ASCO's three categories of cutaneous irAE and is therefore managed apart from inflammatory and bullous dermatoses, what ASCO counts as a SCAR, its presenting symptoms, the AGEP caveat, ASCO's cutaneous timing statement, the temporary-or-permanent discontinuation these manifestations carry, the grade 4 hospitalization statement, and the cross-toxicity grade ladder. The FIRST block is ASCO's cross-organ Recommendations summary, also printed in the abstract, and is cited as the general default rather than as a SCAR-specific recommendation.Supporting text: the full text + the PubMed abstract (checkable at the link above)
  • Haanen J, et al. (2022) Management of toxicities from immunotherapy: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-upCited for: That these entities are rare, that repeated severity evaluation exists precisely to exclude them, and that they carry immediate interruption of ICI therapy with specialist treatment and monitoring; the competence requirement that prescribers recognise early DRESS, Lyell disease and Stevens-Johnson syndrome; the pre-treatment skin and mucosal examination that makes a new lesion interpretable; the referral assessment; and the attribution work a patient on several medications requires.Supporting text: the full text
  • Brahmer JR, et al. (2021) Society for Immunotherapy of Cancer (SITC) clinical practice guideline on immune checkpoint inhibitor-related adverse eventsCited for: That SJS/TEN is rare and life-threatening, has been reported with both classes, is accompanied by mucosal involvement at named sites, and may be rapid or may emerge progressively from a lichenoid eruption that fails usual therapy; the morphological trigger for suspecting it; the biopsy and serology panel that separates it from the bullous dermatoses; the ICI interruption; the consultation triggers; and the infection rule when a rash fails immunosuppression. SITC's steroid-sparing list is deliberately NOT stored here — see this entry's omitted[secondLine].Supporting text: the full text

Educational use only. Educational synthesis of the published literature — not a treatment protocol, dosing guide, or medical advice. Regimens and agents shown are illustrative of what the literature describes; verify against current guidelines (ASON / KDIGO / ASCO / NCCN) and individualize to the patient. Using this site creates no clinician–patient relationship.

Spotted an error? Tell us. Sourcing and method are described on Methods.