Article Text
PDF
Abstract
Ophthalmic immune-related adverse events (Eye-irAEs) from immune checkpoint inhibitors can cause visual morbidity. The absence of standardized definitions for Eye-irAEs not only impedes the development of evidence-based treatments but also progress in translational research. The objective of this study was to develop consensus guidance for an approach to Eye-irAEs.
Four ophthalmic physicians (uveitis specialists and neuro-ophthalmologists) drafted Eye-irAE consensus guidance and definitions, which were reviewed by the multidisciplinary Eye-irAE definition panel. The panel was divided into Group A (Neuro-ophthalmology/Orbital Disease) and Group B (Uveitis/Ocular Surface Disease). A modified Delphi consensus process was used, with two rounds of anonymous ratings by panelists and two meetings to discuss areas of controversy. For each disorder, five diagnostic components were evaluated: symptoms, examination findings, laboratory studies/imaging findings, diagnostic criteria, and treatment. Panelists rated content for usability, appropriateness and accuracy on 9-point scales in electronic surveys and provided free-text comments. Aggregated survey responses were incorporated into revised definitions. Consensus was based on numeric ratings using the RAND Corporation/ University of California Los Angeles Health Services Utilization Study (RAND/UCLA) Appropriateness Method with prespecified definitions.
29 panelists from 25 academic medical centers voted on 114 rating scales (66 neuro-ophthalmic/orbital disease components, 48 uveitis/ocular surface disease components); of these, 86.3% (57/66) in Group A and 89.6% (43/48) in Group B reached first-round consensus. After revisions, all items except 6.1% (4/66) in Group A and 1.6% (1/60) in Group B received second-round consensus. Consensus definitions were achieved for 10/11 neuro-ophthalmic/orbital disorders: optic neuritis, inflammatory optic disc edema, arteritic ischemic optic neuropathy, optic perineuritis, orbital inflammation, thyroid eye disease-like orbital inflammation, cavernous sinus syndrome, oculomotor mononeuritis, trochlear mononeuritis, and abducens mononeuritis. Consensus definitions were achieved for 9/10 uveitis/ocular surface disorders: anterior uveitis, intermediate uveitis, posterior uveitis, panuveitis, Vogt-Koyanagi-Harada-like syndrome, sarcoidosis-like syndrome, acute macular neuroretinopathy, dry eye disease, and scleritis.
These disease definitions establish a standardized classification for Eye-irAE, highlighting differences between irAEs and other inflammatory disorders. Importantly, diagnostic certainty does not always align directly with the need to treat as an Eye-irAE. Given the consensus from this representative panel group, it is anticipated the definitions will be used broadly across clinical and research settings.
- Immune related adverse event - irAE
- Immunotherapy
- Immune Checkpoint Inhibitor
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
Introduction
Immune checkpoint inhibitors (ICIs) are a class of cancer therapy that modifies the patient’s own immune system to better detect and destroy tumor cells. The US Food and Drug Administration has approved many ICIs1 2 for a wide range of malignancies.3 ICIs disrupt immune homeostasis and may lead to exacerbation of autoimmune conditions and new-onset immune-related adverse events (irAEs),4 5 usually occurring at a median of 2–16 weeks from therapy initiation.6
IrAEs may also involve any part of the eye and pathways that govern vision and eye movements and are here defined as Eye-irAE (also described in the literature as ophthalmic irAEs).7–9 There is strong evidence for causality between ICIs and various Eye-irAEs.10–12 While previous case series identified approximately 1% of patients on ICIs developing an Eye-irAE,13 there is a lack of standardized disease definitions. However, for other organ systems, consensus criteria for irAEs have been described, including for cutaneous and neurologic irAEs (irAE-N), which included some irAEs of ophthalmic interest such as myasthenia gravis (MG).14 15
Consensus disease definitions for the diagnosis of and treatment indications for Eye-irAEs are provided here. Additionally, our expert panel offers guidance for managing Eye-irAEs.
Methods
The working group of two neuro-ophthalmology specialists and two uveitis specialists (BKC, KK, ELC, LS) drafted the Eye-irAE guidelines. These included initial disease definitions, signs and symptoms, laboratory and radiological workup, diagnostic criteria, and treatment. This draft was reviewed by a panel of subspecialists using the previously described modified Delphi process.14 16 Surveys of drafted guidelines were sent out for two rounds of voting, commentary, and virtual discussion. The working group used this modified Delphi process to revise the draft into accepted guidelines.
The panel included physicians specializing in cornea, neuro-immunology, neurology, neuro-ophthalmology, oculoplastics and orbital surgery, medical oncology, ocular oncology, and uveitis. Given the diversity of subspecialists, the guidelines were divided into two surveys (Survey A and Survey B). Survey A consisted of guidelines for Eye-irAE diseases pertaining to neuro-ophthalmology and the orbit. Survey B consisted of guidelines for Eye-irAE diseases pertaining to uveitis and the ocular surface. Each panelist chose which survey to participate in given their area of expertize. Participants in Survey A and Survey B underwent two rounds of voting and virtual group meetings to discuss areas of controversy.
During each survey, the panelists anonymously rated drafted guidelines based on usability, appropriateness, and accuracy on a scale of 1–9 with the opportunity to provide free-text comments. Consensus was achieved based on the RAND Corporation/ University of California Los Angeles Health Services Utilization Study (RAND/UCLA) Appropriateness Method if the median survey results fell within the 7–9 point range and fewer than one-third of participants voted outside this range.16 The median ranges were categorized as 1–3 not usable, 4–6 uncertain, and 7–9 usable. If for any part of the draft definition, consensus was reached in the first survey round, it most often did not undergo further revision for the second survey round unless points of non-consensus were raised. The Delphi process was exempted by the Massachusetts General Brigham Institutional Review Board (Protocol #2020P003032).
Results
34 physicians were invited to participate and 29 accepted. They represented 25 academic centers from the USA (22), Canada (1), Australia (1) and Israel (1). Of the 29 participants, 29 (100%) completed the Round 1 survey and 22 (75%) completed the Round 2 survey. The Round 1 panel included neuro-ophthalmologists (12), uveitis, cornea, and oculoplastics specialists (13), oncologists (2), and neuroimmunologists with irAE expertize (2). Seven Round 1 participants did not participate in Round 2. The participant breakdowns for Rounds 1 and 2 are shown in online supplemental figures 1 and 2, respectively.
Supplemental material
The panel identified the following unmet needs for Eye-irAE disease guidelines (% of panel members identifying the issue): (1) identifying subclinical or mild disease (76%); (2) recognizing the spectrum of presentations (83%); (3) differentiating Eye-irAEs from alternative etiologies (97%); (4) grading Eye-irAE severity (76%); (5) classification of patients for cohort studies (55%); (6) classification of Eye-irAE phenotypes for translational research (55%); and (7) adjudication of Eye-irAEs in clinical trials (79%). Consensus statements and definitions were developed with the goal of fulfilling unmet needs in these areas.
Round 1 included the rating of six components repeated for 11 diseases in Group A and 8 diseases in Group B (online supplemental table 1a). 16 individuals completed Survey A, and 86.4% (57/66) of items reached first-round consensus. 13 members completed Survey B, and 89.6% (43/48) of items reached first-round consensus. The Group B panel recommended adding two additional Eye-irAE syndromes: irScleritis and irEpiscleritis. Disease definitions for these syndromes were incorporated into the second round of surveys.
Supplemental material
Round 2 included ratings of 12 revised components in Survey A (the 9 components that did not previously meet consensus and the 3 additional components in orbital myositis that previously met consensus but for which there was sufficient controversy in the virtual meeting that re-voting was deemed necessary). Additionally, it included ratings of 8 revised components (the 5 components that previously did not meet consensus and 3 additional components in immune-related dry eye disease that previously met consensus but for which there was sufficient controversy in the virtual meeting that re-voting was deemed necessary) and 12 new components under 2 new syndromes in Survey B (online supplemental table 1b). The overall retention rate for Round 2 was 75.86% (22/29). In Group A, 13 out of 16 original members participated resulting in a retention rate of 81.25%. In Group B, 9 out of 13 original members participated resulting in a retention rate of 69.23%. In Group A, 93.9% (62/66) reached final consensus. In Group B, 98.3% (59/60) reached final consensus.
Consensus recommendations
Approach to Eye-irAE diagnosis and general guidance statements
Eye-irAEs were broadly classified anatomically into ocular surface disorders, uveitis, orbital disorders and neuro-ophthalmic diseases. Note that the neuro-ophthalmic irAEs discussed in this paper primarily affect the optic nerve and other cranial nerves (CNs). In contrast, the neurologic irAEs discussed in the irAE-N guidelines primarily affect the peripheral nervous system (PNS) and central nervous system (CNS), but may have some overlapping neuro-ophthalmic implications (eg, immune-related neuromuscular junction disorders).14
Each Eye-irAE guidance statement describes: possible symptoms and signs, supportive examination findings, recommended tests, diagnostic criteria, and treatment recommendations. These descriptions are intended to be dynamic: as the diagnostic workup evolves and the patient undergoes appropriate evaluation by the appropriate specialist, the diagnosis may be updated. For instance, an Eye-irAE may first be described as uveitis, and then the subtype of uveitis may be delineated.
Several features increase the likelihood that visual symptoms or signs starting after ICI administration represent an Eye-irAE. The diagnosis requires that other potential etiologies have been reasonably excluded through appropriate workup. Most Eye-irAEs occur early in treatment, usually within 6 months of starting ICIs, although they could occur anytime while on treatment or after completion.17 However, generally to be considered an Eye-irAE, symptoms should begin within 12 months of the last ICI dose. If a new condition occurs 6–12 months or later after the last ICI dose, diagnosis of an Eye-irAE may require a higher burden of proof than those outlined in these guidance statements.
While improvement in symptoms and signs with holding ICI therapy and/or initiation of corticosteroids or other immunomodulators is not specific, this occurrence increases diagnostic certainty of an Eye-irAE. However, some Eye-irAEs may be treatment resistant, result in chronic disease, and/or have irreversible deficits. Therefore, lack of improvement with appropriate treatment may prompt an expansion of the differential diagnosis but does not exclude the possibility of an Eye-irAE. Some visual symptoms such as blurred vision without objective ocular abnormalities may be too non-specific to assign a particular Eye-irAE syndrome or could be related to other irAEs, for example, CNS disease, or to non-irAE complications of cancer therapy (eg, radiation optic neuropathy).
Eye-irAE consensus definitions
For each section, possible symptoms were proposed followed by supportive examination findings. Ancillary tests including laboratory tests and imaging, were divided into required tests, recommended tests and other suggested workup. Based on the symptoms, physical examination, and required ancillary tests, the definite definition was proposed. Probable and possible disease definitions based on all findings including recommended and suggested tests are provided. The treatment indications and recommendations about holding ICIs are also provided.
Diagnostic and management features for Panel A and B diseases are summarized in tables 1 and 2. The online supplemental tables 2 – 22 provide the complete Delphi process findings and should be the primary source of reference for investigators and clinicians.
Panel A: optic nerve and orbital disorders
Panel B: uveitis and ocular surface disease
Panel A: optic nerve and orbital disorders
Optic nerve disorders
Immune-related optic neuritis
Common: Patients developing optic neuritis (ON) can present with either unilateral or bilateral findings and symptoms, which generally include an acute loss of vision that is commonly associated with pain with eye movements. Signs of optic neuropathy should be present (diminished visual acuity and color vision, an afferent pupillary defect in unilateral disease, and possibly optic disc abnormalities such as edema and/or pallor). It is recommended to obtain visual field (VF) tests early on to aid with diagnosis and serve as a baseline for comparison post-treatment (online supplemental table 2).
MRI of the brain and orbits with and without contrast is required for diagnosis. T1 post-contrast fat-suppressed sequences and T2/short tau inversion recovery sequences are more sensitive for the detection of ON-related signal changes. An MRI brain assists in detecting a more disseminated CNS process.
While these typically do not closely mimic inflammatory optic neuritis, infectious optic neuropathies including syphilis, varicella zoster virus (VZV), Lyme and tuberculosis (TB) should be ruled out, especially if risk factors are present.
A definite diagnosis of immune-related ON (irON) can be established if there is a compatible clinical history, examination, and MRI finding of optic nerve enhancement anterior to the optic chiasm. Serum antibodies for other etiologies of ON, including neuromyelitis optica spectrum disorder (NMOSD) and myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD) should be tested. The expert panel did consider as supportive features of irON the presence of AQP4 and MOG antibodies, and markers of multiple sclerosis, such as demyelinating plaques and positive oligoclonal bands. While these are indubitably markers that would support a diagnosis of spontaneous optic neuritis, it is less clear whether this equally applies to irON. For instance, a recent study showed that worsening of MS is likely low.18 Therefore, in the presence of NMOSD or MOGAD antibodies or a clinical picture suggestive of MS, it is currently not clear how to distinguish between an irON, spontaneous ON or paraneoplastic ON as has been described with non-ICI NMOSD.
IrON may present atypically when compared with spontaneous ON. Typical spontaneous ON is most frequently unilateral and associated with ocular pain or pain with eye movements and less commonly with optic disc edema. IrON is often bilateral and painless and tends to present with variable VF defects that differ from the classic central scotoma. Dyschromatopsia is less commonly present. Optic disc edema is more likely to be seen.19 20
Most cases of irON should be treated, including withholding of ICI and initiation of immunomodulation. Intravenous methylprednisolone (IVMP) at a dose of 1000 mg daily or equivalent should be initiated as soon as infectious causes are ruled out and can be continued for 3–5 days. Further, the treatment should be titrated based on the severity of vision loss and response to corticosteroid treatment. Subtypes, evaluations, and definitions for ON are in online supplemental table 2. For some etiologies of spontaneously occurring ON such as MOGAD, an oral steroid taper is needed, although the exact taper duration is not well established for irON. If response to corticosteroids is insufficient, escalated treatment with plasmapheresis can be considered.
Possible: Other laboratory tests, lumbar puncture (LP) and MRI of the spinal cord may be necessary when the physical examination and MRI findings are atypical for isolated ON. When the enhancement pattern is perineural, one should especially consider LP to rule out leptomeningeal disease from the primary malignancy. Optical coherence tomography (OCT) and rarely, visual evoked potentials can be used to confirm and assess the severity of optic neuropathy.
Immune-related inflammatory optic disc edema
Common: Patients with optic disc edema (ODE) may present with unilateral or bilateral findings. Historically, inflammatory ODE has sometimes been referred to as papillitis, but this terminology is poorly defined and misleading. Therefore, the expert committee adjudicated that the term immune-related inflammatory ODE (irODE) is preferable. In the development of these guidelines, a decision was made to assign irODE (with inflammation restricted to the optic nerve head within the eye) to a category separate from irON (see Immune-related optic neuritis sectionabove), in which more extensive optic nerve inflammation is present retro-orbitally that in most cases is appreciable on orbit MRI. Thus, irODE is a reported Eye-irAE that stands at the intersection between intraocular inflammation (ie, uveitis) and retrobulbar inflammation (ie, ON) and the diagnosis should only be made when other signs of uveitis or ON are not present (online supplemental table 3).
Especially if optic disc swelling is bilateral, papilledema due to increased intracranial pressure (ICP) must be ruled out. In addition, cerebrospinal fluid (CSF) inflammation (ie, meningitis) and infiltrative optic neuropathies (eg, from hematologic malignancies or sarcoidosis) may present with ODE. In papilledema, visual dysfunction is a late occurrence, whereas it is expected to be more prominent and present earlier in inflammatory ODE. Other symptoms of increased ICP should be inquired about, and a low threshold maintained for obtaining an LP with opening pressure measurement. VF loss and disc leakage on fundus fluorescein angiography (FFA) may be additional indicators of ODE. An ocular B-scan ultrasound, fundus autofluorescence (FAF), or enhanced depth OCT can help rule out the presence of occult optic disc drusen (ODD), which can give the appearance of ODE. Another important differential diagnosis is anterior ischemic optic neuropathy (AION), which typically presents with sudden vision loss associated with ODE and hemorrhages. AION can occur spontaneously (non-arteritic anterior ischemic optic neuropathy) or in the setting of a vasculitis, such as giant cell arteritis (GCA) or arteritic anterior ischemic optic neuropathy (AAION), a condition that has also been reported as an Eye-irAE (see Immune-related arteritic ischemic optic neuropathy section below).
A definite diagnosis of irODE can be made in the presence of ODE in either eye or bilaterally when mimics such as papilledema, ODD, and AION have been appropriately ruled out. An MRI should be performed to exclude signs of ON and increased ICP. LP should show a normal CSF opening pressure. Supportive features include the absence of venous thrombosis on MRI venous images and normal laboratory markers for inflammation and infection, acknowledging that these may be elevated unrelated to the ODE in the context of malignancy and immunotherapy.
IrODE may not need to be treated unless there is vision compromise such as a VF defect. In the case of vision compromise, consider treating as irON.
Immune-related arteritic ischemic optic neuropathy
Common: Vasculitis is a rare complication of checkpoint inhibitors, but if present, mostly involves large vessels in the form of GCA or eosinophilic granulomatosis with polyangiitis.21 22 Patients are typically over 50 years old and women are at higher risk.23 The resulting AAION typically presents with precipitous or sometimes stuttering vision loss that can be either unilateral or bilateral. In the setting of arteritis, visual symptoms are commonly accompanied by manifestations of cranial hypoperfusion such as headache, jaw claudication, or scalp tenderness as well as systemic symptoms such as fatigue, fevers, or myalgias. Signs of optic neuropathy should be present, such as diminished visual acuity, dyschromatopsia, an afferent pupillary defect, and VF defects. The optic disc is typically swollen but can occasionally be normal in posterior ischemic optic neuropathy. Choroidal hypoperfusion on FFA is suggestive of vasculitis and inflammatory markers including erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) are commonly elevated but can occasionally be normal. A definite diagnosis can be ascertained through temporal artery biopsy, but diagnostic yield is dependent on the timing, length, and quality of the biopsy. In older patients with typical symptoms and elevated ESR and CRP, the diagnosis is probable and should trigger urgent treatment (online supplemental table 4).
Treatment in acute settings is high dose (1000 mg) IVMP for 3–5 days followed by a very slow oral corticosteroid taper. Subtypes, evaluations, and definitions for AAION are in online supplemental table 4. Given the potential for sudden severe complications of vasculitis, strong consideration should be given to holding ICI treatment.
Possible: On examination, there may be retinal ischemic manifestations. Arteritis may also involve efferent pathways and cause diplopia. Other diagnostic modalities can be helpful in difficult cases such as temporal artery ultrasound, cranial MRI with post-contrast vessel wall imaging of the extracranial circulation, and a vasculitis protocol positron emission tomography-CT of the body. Analogous to spontaneously occurring GCA, treatment with tocilizumab can be considered to avoid adverse effects of long-term corticosteroid use.
Immune-related optic perineuritis
Common: Optic perineuritis is an inflammation of the optic nerve sheath that may cause pain with eye movements and vision loss. Unlike ON, where inflammation of the parenchyma of the optic nerve is primary, isolated optic perineuritis is a disorder of connective tissue inflammation. Patients who develop optic perineuritis present with unilateral or bilateral symptoms. Timing of symptom onset is typically gradual over a period of days, slower than in most cases of ON. The inflammation is typically evident as gadolinium enhancement of the optic nerve sheath on orbital MRI. Demyelinating disorders such as MOGAD and NMOSD may cause optic perineuritis but generally in association with ON. It is recommended to maintain a broad differential diagnosis and check laboratory studies as optic nerve sheath enhancement can be seen in a variety of infectious (VZV, TB), inflammatory (antineutrophilic cytoplasmic antibody (ANCA) vasculitis, immunoglobulin G4 (IgG4)-related disease, GCA) and oncologic (hematologic malignancies, optic nerve sheath meningioma) diseases (online supplemental table 5).
Optic perineuritis typically improves briskly with systemic steroids and rapid response to corticosteroid treatment is supportive of diagnosis.
Possible: If MRI findings and the clinical picture are supportive but the patient does not respond to corticosteroid rapidly and optimally, then the diagnosis remains possible, but other diagnostic considerations should be entertained. In this situation, additional workup may be appropriate, including LP to rule out perineural malignant leptomeningeal disease.
Orbital diseases
Immune-related orbital inflammation
Common: Patients suffering from orbital inflammation (OI)/orbital pseudotumor typically present with unilateral or bilateral eye pain (sometimes exacerbated by ocular movements) and visible signs of orbital inflammation such as periorbital swelling, proptosis, chemosis, lacrimation, conjunctival erythema or tenderness. In immune-related OI (irOI), the inflammation is often more diffuse than in thyroid eye disease, which tends to involve extraocular muscles prominently. Visual dysfunction may be present and can include binocular diplopia and vision loss (implying optic nerve compromise). CT and MRI imaging of the orbit with and without contrast and with fat-suppressed T1/T2 imaging is recommended. Since many infectious, inflammatory, and neoplastic processes can be centered in the orbit, consideration should be given to obtaining a biopsy in order to establish a definite clinicopathologic diagnosis. However, if presentation is acute and typical, an empiric trial of corticosteroids can be attempted first, and a brisk and complete response would also establish the diagnosis (online supplemental table 6).
Obtaining laboratory markers for thyroid eye disease and systemic inflammatory conditions should be considered. In some cases, otolaryngologist evaluation may be appropriate to assess for infection or inflammation of adjacent paranasal sinuses. If a biopsy is performed, evaluation should include stains for infections and immunohistochemical assessment for clonality and IgG4-positive plasma cells to rule out IgG4-related disease. Orbital inflammation is typically treated with moderate to high doses of oral corticosteroids (eg, prednisone 1 mg/kg/day) with a slow taper.
Immune-related thyroid eye disease-like OI
Common: An orbital inflammatory disorder closely mimicking thyroid eye disease (TED) may occur as an irAE. Patients with TED present with unilateral and bilateral symptoms that commonly include eyelid retraction, lid lag, proptosis, periorbital swelling, chemosis, erythema, ocular misalignment, and diplopia. Severe cases may be complicated by vision loss related to compressive optic neuropathy. A history of Grave’s disease and symptoms of hyperthyroidism or hypothyroidism would provide additional support. CT and MRI imaging of the orbit and face with and without contrast can indicate fusiform enlargement of orbital recti muscles, which usually spare the muscle tendons.24 Imaging in more inflammatory myositis typically shows diffuse muscle enlargement. OCT and VF should be used to monitor for possible optic nerve compromise. Testing is recommended for antibodies to the thyroid stimulating hormone receptor, thyroid peroxidase, and thyroglobulin. A definitive diagnosis is made based on TED symptoms, positive thyroid antibodies, and supportive imaging findings. In the presence of TED symptoms and positive thyroid antibodies but without supportive imaging findings, the diagnosis is probable. In most cases, observation without treatment or with limited supportive treatments such as artificial tears are sufficient. Selenium supplementation has been recommended in mild TED. If the patient is a smoker, smoking cessation should be advised (online supplemental table 7).
Possible: If a patient has isolated TED symptoms or imaging findings of enlarged extraocular muscles in the absence of any other more probable diagnosis, then a diagnosis of TED is possible. In cases of severe symptoms or optic nerve compromise, high-dose intravenous or oral steroids should be used.
Uncommon: If there is evidence of compressive optic neuropathy, surgery should be considered. There is evidence for the efficacy of the insulin-like growth factor-1 receptor blocker teprotumumab in more severe cases of TED, but it should be avoided in the presence of a malignancy. Rarely, in severe refractory cases, orbital radiation can be performed.
Immune-related cavernous sinus syndrome
Common: Patients with cavernous sinus syndrome may suffer from unilateral or bilateral ocular movement restriction and binocular double vision related to dysfunction of one or more of CN III, IV, and VI or sympathetic nerves. Additionally, there may be orbital pain and/or facial numbness, symptoms related to dysfunction of the first two branches of the trigeminal nerve (CN V). Other symptoms may include ptosis and pupillary changes (related to CN III involvement or Horner syndrome) and signs of congestive orbitopathy (proptosis, conjunctival injection or chemosis) (online supplemental table 8).
A brain and orbit MRI with and without contrast are recommended. Additional vascular imaging to rule out a carotid-cavernous fistula can be useful. Although the cavernous sinus is outside the subarachnoid space, an LP may be useful to assess for evidence of CSF inflammation and other processes such as hematologic malignancies.
In most cases, a diagnosis can be achieved without a biopsy if appropriate symptoms and signs are present and the process briskly responds to corticosteroids, especially since biopsy of a deep and highly vascular structure such as the cavernous sinus carries considerable risk.
If a diagnosis is made without a biopsy and the patient responds to steroids but has a recurrence, lymphoma should be considered.
Immune-related mononeuritis of cranial nerves III (oculomotor), IV (trochlear), VI (abducens)
Common: Mononeuritis of any one of the three nerves innervating the extraocular muscles typically presents with diplopia. This is horizontal for CN VI, mostly vertical for CN IV and can be a combination of both or be vertical-oblique in the case of CN III. The latter also innervates the pupil and levator palpebrae superioris muscle and, therefore, CN III palsy typically presents with ptosis and ipsilateral mydriasis. Eye pain or headache may also accompany inflammation or infarction of any of these CNs (online supplemental tables 9,10,11).
It should be noted that extraocular muscle abnormalities and binocular diplopia in the setting of ICI use are more typically related to immune-related MG or immune-related myositis than to mononeuritis of CNs. These are defined in the previously published irAE-N guidelines and should be carefully excluded.
In addition to an irAE, alternative causes of cranial neuropathy should be considered, including nerve compression and microvascular disease. Particularly in the case of a CN III palsy, urgent vascular imaging with CT angiography or magnetic resonance angiography of the head should be performed to rule out an enlarging aneurysm. Once this is ruled out, an MRI of the skull base, cavernous sinus, and orbit with contrast and fat-suppressed sequences should be obtained to assess for inflammation anywhere along the course of these nerves. If an inflammatory cranial neuropathy is suspected, additional serologies may provide clues, as outlined. This may include testing for GQ1b antibodies particularly in the presence of multiple cranial neuropathies and other features of the Miller Fisher syndrome such as ataxia or areflexia.
A definitive diagnosis of an Eye-irAE affecting one of these CNs is made based on the presence of classic symptoms of nerve palsy, negative imaging for metastasis or aneurysm, and rapid improvement of symptoms with steroids. Treatment is typically with high-dose oral steroids with slow taper to prevent rebound.
Immune-related orbital myositis
Common: Patients with orbital myositis (OM) typically present with unilateral or bilateral ocular movement restrictions. Symptoms commonly include binocular double vision, painful ophthalmoparesis, ptosis, chemosis, conjunctival injection, and eye pain with movements. Ocular misalignment and extraocular muscle weakness are additional supportive findings. Thus, the symptoms may overlap with irOI and immune-related TED-like OI (online supplemental table 12).
When occurring as an irAE, irOM may overlap with other neuromuscular irAEs including systemic myositis, MG and even myocarditis. Thus, there may also be systemic symptoms, causing muscle pain and weakness such as dysphagia, dysphonia, dyspnea, and chest pain. Skin changes associated with dermatomyositis, periorbital edema, eyelid edema and erythema may also occur. Please refer to the irAE-N consensus statement for guidelines on immune-related MG and systemic myositis.14
MRI or CT imaging of the orbit with and without contrast is recommended to identify enlarged and edematous orbital muscles. A biopsy showing OM could provide a definitive diagnosis and rule out systemic myositis but may not be necessary in many cases; often a definitive diagnosis can be made based on a combination of the classic symptoms and imaging findings in the absence of systemic myositis and MG.
First-line treatment typically consists of high-dose oral corticosteroids ranging 0.5–1.0 mg/kg/day with a slow taper. It is important to continue monitoring for systemic involvement such as respiratory and cardiac compromise.
Panel B: uveitis and ocular surface disease
Immune-related anterior uveitis, immune-related intermediate uveitis, immune-related posterior uveitis, immune-related panuveitis
We have grouped the consensus results for the four anatomic locations of uveitis (anterior, intermediate, posterior and panuveitis) together here as there is significant overlap between them. We have also highlighted differences between them, particularly with regard to treatment (online supplemental tables 13–16).
Common: Patients who develop uveitis (immune-related anterior uveitis (irAU), immune-related intermediate uveitis (irIU), immune-related posterior uveitis (irPosU), immune-related panuveitis (irPanU)) typically present with bilateral findings and symptoms, although cases can be asymmetric. As with other irAEs, onset is typically within the first 6 months after initiation of ICI therapy, although late presentations can occur. Symptom onset is generally acute. For irAU, the most common symptoms are eye redness, eye pain, headache, photophobia and blurry vision. For irIU and irPosU, floaters and photopsias are more frequently noted than in irAU, but they can also have irAU symptoms. Because the entire eye is inflamed in irPanU, any of the above symptoms can be noted. Diagnostic findings include inflammatory cells in the eye compartment indicated by the disease. For irAU, there is anterior chamber cell and/or flare on slit lamp examination. For irIU, vitreous cells or haze are present as well as pars plana vitreous exudates. For irPosU, inflammatory chorioretinal lesions and/or retinal perivascular exudates are present. Vitreous cells or haze are also often present. For irPanU, supportive examination findings include the presence of anterior chamber cell or flare in addition to inflammation in the vitreous, retina, and/or choroid. Anatomic complications of uveitis may be present on examination. Cystoid macular edema (CME) can occur with any anatomic location of uveitis. Choroidal neovascular membrane (CNVM) typically is only found with irPosU and irPanU.
For all uveitis anatomic locations, testing is recommended to rule out other etiologies of uveitis, particularly infectious etiologies. This includes serum testing and imaging as outlined in online supplemental tables 13–16. For all uveitis anatomic locations, testing to rule out syphilis, tuberculosis, Lyme disease (in endemic areas) and sarcoidosis should be done. For irAU, human leukocyte antigen B27 (HLA-B27) should additionally be checked. For irIU, an MRI brain to evaluate for demyelinating disease should additionally be checked if the patient has neurologic symptoms. For irPosU and irPanU, if retinal vasculitis is also present, testing for lupus, ANCA vasculitis and IgG4 disease should additionally be checked.
For irAU, a topical ophthalmic corticosteroid is recommended with follow-up every 1–2 weeks until symptoms and signs resolve. The corticosteroid regimen should be clinically adjusted to the lowest possible dose that maintains quiet disease. If posterior synechiae develop, the addition of a topical cycloplegic agent is needed. Consensus was reached that the ICI should not be discontinued and that irAU can generally be successfully controlled with topical therapy.
For irIU, we recommended initiating a topical ophthalmic corticosteroid with follow-up every 1–2 weeks. If findings persist on follow-up, then periocular or intraocular corticosteroid injections may be considered.
For irPosU, if the disease is mild, it could be observed with close follow-up. If findings persist or worsen on follow-up, then intraocular steroids may be considered. If the disease progresses despite this, consider holding the ICI and/or initiating oral corticosteroids.
For irPanU, the recommended treatment would be to initiate a topical ophthalmic corticosteroid to treat the anterior segment inflammation with follow-up every 1–2 weeks. If findings persist or worsen on follow-up, then intraocular steroids may be considered. If disease progresses despite this, consider holding the ICI and/or initiating oral corticosteroids.
Possible
If the diagnosis of irAU is uncertain (eg, inflammation may not be isolated to the anterior chamber), then additional ancillary testing can be considered including FAF, FFA, and macula OCT. This would help rule out other immune-related uveitic diseases. For treatment, if the patient is unresponsive to topical corticosteroids, periocular or intraocular injectable corticosteroids may be considered.
If the diagnosis of irIU is uncertain, then additional ancillary testing can be considered including FFA and macula OCT. As with irAU, this would help rule out other immune-related uveitic diseases. For treatment, if CME is detected on ancillary testing, topical non-steroidal anti-inflammatory drugs (NSAIDs) may be added with further escalation to intraocular steroids as necessary. If vascular leakage is detected on FFA, escalation to intraocular steroids may be necessary.
If the diagnoses of irPosU or irPanU are uncertain, then additional ancillary testing can be considered including FAF, FFA, indocyanine green angiography (ICGA), macula OCT, and OCT angiography. As with irAU and irIU, this would help rule out other immune-related uveitic diseases. These tests could also detect the presence of focal chorioretinitis, CME, CNVM, and retinal vasculitis. As with irIU, if CME is detected on ancillary testing, topical NSAIDs may be added with further escalation to intraocular steroids as necessary. If vascular leakage is detected on FFA, escalation to intraocular steroids may be necessary. If CNVM is detected on FFA or OCT angiography, intravitreal anti-vascular endothelial growth factor injections may be necessary.
Uncommon: If any of the uveitis anatomic locations cannot be adequately controlled with localized corticosteroid therapy, then a scheduled break in ICI may be discussed with the oncology team. In addition, short-term systemic corticosteroid therapy and long-term immunomodulatory therapy may be discussed with the oncology team.
Immune-related Vogt-Koyanagi-Harada-like syndrome
Common: Immune-related Vogt-Koyanagi-Harada-like syndrome (irVKHLS) is a specific type of irPanU characterized by diffuse choroidal thickening and serous retinal detachments. Patients who are developing irVKHLS typically present with bilateral findings and symptoms weeks to months following initiation of ICI therapy. The most common symptoms are blurred vision, red eye, eye pain, and floaters. Headache, hearing loss, vitiligo, and poliosis are usually less common than in spontaneous VKH but may also be present. Examination findings can include the presence of anterior chamber cell or flare, mutton fat keratic precipitates, and vitreous cells and/or haze. Multiple serous retinal detachments are the finding that is most suggestive of irVKHLS in the presence of other findings consistent with irPanU. Optic nerve hyperemia and/or disc edema may also be present. The same testing to rule out other etiologies of irPanU as in online supplemental table 16 should be done. Treatment with topical, periocular and/or intraocular corticosteroids, as described above for irPanU, are the typical treatment approach (online supplemental table 17).
Possible: If the diagnosis of irVKHLS is uncertain, and if ancillary testing has not been obtained, then it should include: macula OCT, FAF, FFA and ICGA. Findings that would support the diagnosis include large serous retinal detachments with subretinal fluid on OCT macula. FFA may show multiple hyperfluorescent spots with late leakage and pooling as well as optic disc hyperfluorescence. ICGA may show multiple hypofluorescent dark spots.
Uncommon: If the irVKHLS cannot be adequately controlled with localized corticosteroid therapy, then a scheduled break in ICI may be discussed with the oncology team. In addition, short-term systemic corticosteroid therapy and long-term immunomodulatory therapy may be considered.
Immune-related sarcoidosis-like syndrome
Common: Immune-related sarcoidosis-like syndrome (irSLS) can present as any anatomic location of subtype uveitis along with systemic findings to suggest sarcoidosis. Consequently, the signs and symptoms of irSLS depend on the anatomic location of uveitis as delineated above. Concurrent systemic findings may be present including lymphadenopathy, chest pain, shortness of breath, chronic cough, and skin lesions. Suggestive examination findings include the presence of anterior chamber cell or flare, mutton fat keratic precipitates, iris granulomas, and vitreous cells and/or haze. Other supportive findings include choroidal lesions and perivascular exudates. Testing to rule out other causes of uveitis should be done as noted above depending on the uveitis anatomic location. Recommended treatment depends on the uveitis anatomic location as delineated above (online supplemental table 18).
Possible: If the diagnosis of irSLS is uncertain, then additional ancillary testing can be considered including macula OCT, FAF, and FFA. A finding that would be highly suggestive of irSLS is a choroidal granuloma. FFA may show vascular leakage. If possible, the most easily accessible organ should be biopsied to detect the presence of non-caseating giant cell granulomas, which would definitively diagnose sarcoidosis. Diagnostic vitrectomy and tissue biopsy to rule out other causes of panuveitis may rarely be necessary.
Uncommon: If the irSLS cannot be adequately controlled with localized corticosteroid therapy, then a scheduled break in ICI, short-term systemic corticosteroid therapy and long-term immunomodulatory therapy may be discussed with the oncology team.
PD-L1 inhibitor associated acute macular neuroretinopathy
Common: Patients who are developing acute macular neuroretinopathy (AMNR) can present with bilateral or unilateral symptoms. Symptom onset has been most frequently reported within 2 weeks of ICI initiation. Symptoms are usually acute in onset and include blurry vision, scotomas, photopsias, and metamorphopsia. Patients with this condition sometimes develop concurrent systemic findings such as fever, fatigue, malaise, and arthralgia. Supportive examination findings include the presence of petaloid macular lesions and perivascular exudates. No additional testing is needed when classic symptoms are present and the patient is concurrently using programmed death-ligand 1 (PD-L1) inhibitors. Recommended treatment would be observation with follow-up every 1–2 weeks until symptoms resolve. If findings persist on follow-up, then short-term systemic steroids may be considered, particularly if there is a prominent retinal vasculitis (online supplemental table 19).
Possible: If the diagnosis of PD-L1 inhibitor associated AMNR (PD-L1 AMNR) is uncertain, then additional ancillary testing can be considered including near-infrared fundus photos, macula OCT, and FFA. Findings suggestive of AMNR are wedge-shaped lesions on near-infrared fundus photos and vascular leakage on FFA. Macula OCT will show bilateral hyperreflective band-like lesions in the outer plexiform and outer nuclear layers with disruption of the ellipsoid zone. If there are concurrent CNS symptoms, a brain MRI should also be considered as T2-weighted signal in the caudate heads and anterior lentiform nuclei has been reported in patients with concurrent confusion.25
Uncommon: If the PD-L1 AMNR does not resolve with time, then a scheduled break in ICI therapy may be considered.
Immune-related dry eye disease
Common: Patients who are developing immune-related dry eye disease (irDED) typically present with bilateral symptoms. As with uveitis, onset is usually within the first 6 months of ICI initiation. Symptoms include blurred vision, subjective feelings of dryness, foreign body sensation, tearing, photophobia, pain, and eye redness. Patients may endorse that symptoms improve with blinking and worsen with reading or screen use. Lack of prior documentation of DED and related conditions can be key to indicate that symptoms are new in onset and more likely related to ICI initiation. However, since many patients have symptoms of ocular surface disease at baseline, symptoms of increased severity after ICI initiation are also highly suggestive of irDED. Supportive examination findings include meibomian gland congestion, conjunctival injection, and superficial punctate keratopathy. Additional supportive testing includes tear breakup time less than 10 s, Schirmer’s test without anesthetic less than 10 mm, staining of the ocular surface with diagnostic dyes, and elevated matrix metalloproteinase-9 in the tear film. Recommended treatments could include use of topical lubricating eye drops, punctal occlusion, anti-inflammatory eye drops, serum tears, varenicline nasal spray, hydroxypropyl cellulose insert, scleral lenses, and partial tarsorrhaphy. Most cases of irDED may be followed every 1–2 months until symptoms resolve, although clinical discretion is necessary (online supplemental table 20).
Possible: If the diagnosis of irDED is uncertain, then blood tests may be considered to rule out Sjögren’s syndrome. These include SS-A (Ro), SS-B (La), and antinuclear antibody (ANA). However, most patients with irDED will not have positive serologies.
Uncommon: If the irDED cannot be adequately controlled with local therapies and there is risk of permanent damage to the cornea, then a scheduled break in ICI may be discussed with the oncology team. For most cases, ICI may be continued.
Immune-related scleritis
Common: Patients who are developing immune-related scleritis (irScl) typically present acutely with eye injection, photophobia, and deep pain exacerbated by touch. Supportive examination findings include deep scleral injection and edema, tenderness to palpation over the area of injection, failure of vessel blanching with instillation of phenylephrine, and absence of inflammation inside the eye on slit lamp examination. Areas of scleral thinning may be seen. Recommended treatment includes oral NSAIDs or oral systemic corticosteroids titrated to signs and symptoms (online supplemental table 21).
Possible: If the diagnosis of irScl is uncertain, then blood tests may be considered to rule out other causes of scleritis. These include testing for ANCA, ANA, rheumatoid factor (RF), cyclic citrullinated peptide (CCP), and HLA-B27.
Uncommon: If the irScl cannot be adequately controlled with the aforementioned therapies, then a scheduled break in ICI may be discussed with the oncology team. In addition, long-term immunomodulatory therapy may be discussed with the oncology team.
Immune-related episcleritis
Common: Patients who are developing immune-related episcleritis (irEpi) typically have an acute onset of eye injection, foreign body sensation, and/or blurred vision. Supportive examination findings include superficial episcleral injection, presence of vessel blanching with instillation of phenylephrine, and absence of inflammation inside the eye. Blood work is not required for typical cases. Recommended treatment includes lubricating eye drops, topical corticosteroids, and oral NSAIDs (online supplemental table 22).
Possible: If the diagnosis of irEpi is uncertain, then blood work and imaging may be considered to rule out causes of scleritis including ANCA, ANA, RF, CCP, and HLA-B27.
Uncommon: If irEpi appears atypical with findings of nodularity, recurrence, and resistance to conservative treatment, then the diagnosis may be more consistent with irScl, and diagnosis and treatment of irScl may be considered.
Discussion
This multidisciplinary, multi-institutional group has developed consensus disease definitions, diagnosis, and treatment guidelines for Eye-irAEs following ICIs. Given the diversity of group members and the broad spectrum of Eye-irAEs, the definitions were divided into two groups to match each panelist’s area of expertize with the appropriate diseases. For example, neuro-ophthalmologists were prompted to participate in Group A. Although this resulted in smaller-sized subgroups of panelists, it maximized utilization of each panelist’s specific expertize, encouraged participation and minimized attrition.
The original set of guidelines used in this study was drafted from a rigorous process of comprehensive literature review, existing guidance from major oncologic organizations, and opinions of experts with clinical experience of irAEs from various disciplines. Previously published ophthalmic irAE guidelines have been published from the Society for Immunotherapy of Cancer, American Society of Clinical Oncology, European Society for Medical Oncology, and 2022 National Comprehensive Cancer Network.7–9 The 2022 National Comprehensive Cancer Network guidelines provided recommendations for the management of ocular inflammation, orbital inflammation/orbitopathy, retinal/choroidal disease, and optic neuropathy.26 The guidelines set forth here for Eye-irAEs offer more comprehensive diagnostic criteria and treatment recommendations for a greater range of ophthalmic and neuro-ophthalmic toxicities.
Study strengths include high response rate, multidisciplinary panel, detailed comments from panelists, high rates of final consensus, and rigorous methods using the Delphi consensus approach. Given the high rates of final consensus, these guidelines will serve as a helpful tool in clinical practice and research to help guide ophthalmologists, oncologists, and other subspecialists.
One limitation of the study is the retention rate for both rounds of voting. Group A had a retention rate of 81.25% (13/16). Group B had a retention rate of 69.23% (9/13). This may be related to the small size of an expert panel and the accompanying higher variance as well as the diversity in geographic location and therefore time zones of the panelists, which made coordination and participation more challenging.
The Delphi process revealed areas of controversy that were discussed during the virtual panel meetings. For Group A, this included the naming of papilledema versus papillitis versus optic disc edema. Most people agreed that papilledema and papillitis were not interchangeable, but some members suggested papillitis be abolished altogether and instead use more granular terms such as papillitis/uveitis or papillitis/retinitis. Ultimately, “inflammatory optic disc edema” was chosen as the name for the complication and thus was renamed for Round 2 of voting. There was also extensive discussion regarding the necessity of certain tests for cranial neuropathies. Some physicians believed that testing for such exhaustive lists of conditions may result in false positives, be costly, and have a low yield. The more specialized tests such as LP and muscle-specific kinase antibody that are designed to test for specific manifestations of cranial neuropathies have been listed separately as possible additional tests in case the diagnosis is unclear. Furthermore, there was controversy in the treatment components for the disease definitions. This is due to the lack of standardization of treatment in these new ICI-induced diseases. Panelists differed in whether or not they would choose to observe, use intravenous or oral corticosteroids, and when to consider immunomodulatory treatment (IMT).
Another main area of controversy from Group A was in the disease definition irOM. Only two out of six components of the disease definition met final consensus. Some panelists did not want to include systemic symptoms and findings of myositis since the definition is specifically for orbital myositis. However, the authors believe it is necessary to include systemic findings as irOM may be an isolated phenomenon or seen in conjunction with systemic immune-related myositis.27 28 Another reason for the controversy involved the necessity of MG workup in irOM. Some panelists felt that MG testing would be unnecessary. However, the authors believe that it would be prudent to consider MG testing since there is often an overlap between irOM and immune-related MG.29–31
For Group B, all criteria met final consensus by the modified Delphi criteria except for the definition of irEpi. Panelists felt that irEpi may be difficult to distinguish from irScl. One clinical way to distinguish was suggested that if there is rosacea, allergy, or environmental pollution, then irEpi is more likely. Another panelist commented to treat irEpi as irScl if there is no response to initial treatment. Panelists also suggested that irEpi may be difficult to distinguish from irDED.
There were other areas of controversy in criteria, but these did meet consensus based on the modified Delphi criteria. One of these areas was the indication for sending sarcoidosis laboratory studies in patients who present with irScl. Most of the group agreed with ordering such laboratory studies and imaging, but some members feel that the utility is low as sarcoidosis is rarely a cause of scleritis. We have included serologic workup and imaging in our guidelines given the consensus among panelists and their utilization in the International Workshop on Ocular Sarcoidosis.32 However, the utility of these tests remains an area for further research. There was also debate about whether or not anterior chamber cell may be present in irScl. Absence of anterior chamber cell on examination in irScl reached consensus by the Delphi process. However, a few panelists commented that a small amount of anterior chamber cell may be present in irScl. In these situations, experts may disagree on defining a clinical picture as irScl alone versus irScl and irAU. For irSLS, there was also discussion about whether or not to send sarcoidosis laboratory studies. Most panelists agreed to send for ACE and lysozyme. Some panelists suggested that given the lack of a specific biomarker, looking for systemic findings in sarcoidosis may be helpful, especially since the definitive criterion of biopsy may not always be achievable. Also of note, the complication originally named “Acute macular neuroretinopathy associated with atezolizumab” was broadened to “Programmed death-ligand 1 (PD-L1) associated acute macular neuroretinopathy” in Round 2.
Overall, there was some disagreement with the section of “Definition – Diagnostic Criteria.” Most of the commentary was regarding terminology of “classic symptoms” and “classic findings”. The intention was for these terms to refer to items summarized in other parts of the table. Another common concern was that the “probable” and “possible” categories were non-specific and may lead to incorrect diagnoses. The intention of using this system of categorization was to make it accessible for all providers, especially those without all the available diagnostic testing and imaging modalities. Another reason for keeping this system was to maintain consistency with the published irAE-N and dermatologic-irAE-consensus disease definitions created by the respective neurology and dermatology consensus panels.11 12 For all the Eye-irAEs, the panelists agreed there is strong evidence for direct association/causality of ICIs in their development as supported by the literature,10–12 but it is still of paramount importance to always rule out alternative causes, particularly infections and primary malignancy, as delineated in the tables.
ICI-associated autoimmune retinopathy (AIR) was not included in our disease definitions and guidelines given the paucity of these cases. Cases of new onset or worsening retinopathy with positive antiretinal antibodies following initiation of ICI (either ipilimumab, nivolumab, or combination of both) have been reported.33–35 As use of ICI increases and more cases of ICI-associated AIR are identified, this topic can be revisited.
While the panels broadly discussed the degree of vision loss (eg, severe or not) and pattern of vision loss (acute or other), specific thresholds of visual acuity loss in terms of severity and duration were not discussed in the panels. In clinical practice, vision-threatening disease is often a subjective assessment based on the current rate of decline in vision and the severity of the ocular pathology. Medical oncologists on the panel stated that if vision-threatening disease is present, then systemic corticosteroids may be used especially in acute settings. Use of systemic corticosteroids has been supported in irAEs involving other organ systems.36 37 Higher doses of systemic corticosteroids with rapid taper are preferred to prolonged use of lower doses (10 mg or less daily). The timing of when to initiate systemic corticosteroids remains an area of controversy that requires additional research and discussion. Some panelists thought that the phrase “as a last resort” to indicate when steroid or IMT initiation and holding ICI treatment should occur was overly conservative. Some felt these should happen earlier than “as a last resort” if there were severe symptoms or uncontrolled inflammation after less aggressive therapies. The panelists did not discuss any absolute ophthalmic indications for stopping ICI treatment and felt that this decision would always require a multidisciplinary discussion among the oncologist and ophthalmologist, considering the totality of the patient’s clinical situation. Panelists revealed that IMT may be considered in conjunction with close multidisciplinary follow-up. There is limited data on outcomes with second-line immunosuppression for refractory irAEs and it was not explored with this expert panel.38 39
In conclusion, developing consensus irAE definitions for ophthalmic conditions not only helps the field but also promises to streamline patient care, foster research collaboration, and provide a robust framework for accurate diagnosis and treatment across diverse ophthalmic conditions, serving as a crucial resource for multi-disciplinary patient care.
Ethics statements
Patient consent for publication
Ethics approval
This study was indicated as exempt by the Massachusetts General Brigham Institutional Review Board (Protocol #2020P003032).
Acknowledgments
Assistance in figures preparation and data collection was provided by Renisha Jones, Teilo Schaller, and Bill Louv of Project Data Sphere.
References
Footnotes
LS and BKC contributed equally.
ELC and RL contributed equally.
Contributors ELC, KK, and RL contributed equally to this work as co-first authors, with the addition that LS, BKC, and KLR conceptualized and designed the study and KH, JMD, JR, ELC, KK, RL, LS, BKC, and KLR designed, analyzed and interpreted the Delphi surveys and data. All authors provided (1) substantial contributions to the conception or design of the work, or the acquisition, analysis or interpretation of data; (2) drafting the work or revising it critically for important intellectual content; (3) final approval of the version published; (4) agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. MB, MTB, JJC, JC, FC, LAD, LBD, MD, RE, CLF, SKF, SG, LKG, ACG, DBJ, NK, MK, AGL, ML, A-ML-C, DM, GP, MP, JP, KMR, HNS, AS, PSS, ET, and MKY contributed equally as part of the Eye-irAE disease definition panel. LS is the guarantor for this manuscript.
Funding Partial funding for this work was supported by Project Data Sphere (award number: N/A).
Competing interests FC is on the advisory board of Healio Live, Horizon Therapeutics, Sanofi, and Alexion. LAD is a consultant for IDEAYA Biosciences and has funding from the CTSA Grant Number KL2 TR002379 from the National Center for Advancing Translational Science. SKF is a consultant/advisory board member for Amgen, Sling, Viridian, Immunovant, Poriferous, Lassen, Merida/Third Rock, WL Gore, Kriya, Medtronic, Janssen. ACG receives grant support from MGNet Rare Disease Network for the ADAPT teleMG study on telemedicine in MG, Explore MG2 natural history study (NIH Rare Disease Clinical Research Network Consortium supported by U54 NS115054), Wearable Sensor and Digital Technologies for Quantitative Assessment and Remote Monitoring of Symptoms in Myasthenia Gravis (Project Number1R44NS122672), and Project Data Sphere and the Dysimmune Disease Foundation for work in neurologic immune-related adverse events. She is also on the medical advisory boards for UCB, Alexion, Argenx, Regeneron. DBJ has served on advisory boards or as a consultant for AstraZeneca, BMS, The Jackson Laboratory, Merck, Mosaic ImmunoEngineering, Novartis, Pfizer, Targovax, and Teiko, and has received research funding from BMS and Incyte, and has patents pending for use of MHC-II as a biomarker for immune checkpoint inhibitor response, and abatacept as treatment for immune-related adverse events. AGL is a consultant for Amgen, Alexion, Viridian, AstraZeneca, Bristol Myers Squibb, Stoke, NFL, NASA, and US DOJ. HNS is an employee of Sanofi Pharmaceuticals R&D. AS is a speaker for Novartis and Merck and a consultant for Merck and BMS. The remaining authors have no reported competing interests.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.