Abstract
Introduction
The aims of the study were to investigate clinical characteristics and spectrum of immune checkpoint inhibitor optic nerve disorders (ICI-OND), including ICI-optic neuritis (ICI-ON), ICI-optic disc edema (ICI-OE), ICI-optic perineuritis, and ICI-panuveitis; to understand the breadth of management options; to investigate long-term outcomes of these patients; and, importantly, to tackle the question of outcomes following rechallenge of these potentially life-preserving drug regimens.
Methods
Retrospective analysis, single-center tertiary cancer hospital. Thirty-seven eyes in twenty consecutive patients with ICI-OND treated at Memorial Sloan Kettering Cancer Center between June 16, 2019, and June 5, 2025, were included.
Results
In 37 eyes with ICI-OND, the majority of patients had painless loss of vision (85%), bilateral involvement (70%), dyschromatopsia (59%), abnormal testing: visual field (79%), retinal nerve fiber layer optical coherence tomography (80%), and magnetic resonance imaging (56%). Time to diagnosis occurred following a median of 4.5 drug cycles and was shorter with combination ICI (two ICI’s) versus monotherapy (one ICI). The 3 patients with pain all had concurrent intraocular inflammation. Following steroid treatment and ICI cessation, vision was improved in 84% of eyes, stable in 11%, and declined in 5%. Snellen lines regained was associated with baseline vision (p = 0.0002). Five patients were rechallenged, and we did not see nor detect recurrence of ICI-OND in 4 patients; 1 patient had recurrence which resolved with steroids and vision returned to baseline.
Conclusions
In the majority of cases, ICI-OND occurs bilaterally with painless vision decline and color vision deficits. Cessation of drug along with initiation of systemic steroids can improve vision to 20/40 or better in at least one involved eye, and although we included a smaller subset of patients, we found that ICI can be rechallenged after completion of steroids, without documented recurrence of ICI-OND in most patients.
Keywords: Immunotherapy, Checkpoint inhibitor, Uveitis, Optic nerve, Inflammation
Introduction
Immune checkpoint inhibitors (ICIs) are a class of anti-cancer drugs that treat various types of cancer and work by potentiating the immune system to attack cancer cells. The main classes of FDA approved ICIs include cytotoxic T-lymphocyte associated antigen 4, programmed cell death, programmed cell death protein 1, and lymphocyte activation gene 3 inhibitors [1]. This inflammatory response can also be directed against any layer of the eye, including the optic nerve [2, 3]. This retrospective case series describes a cohort of patients diagnosed with ICI-optic nerve disorders (ICI-OND) and their most common clinical features, how they were managed, and their long-term outcomes including after rechallenge. The balance between vision loss from drug toxicity and the potential preservation of life from ICI makes the question of drug rechallenge particularly pertinent.
Methods
This study includes and provides expanded data on 7 patients (from a multicenter cohort of 11 patients) that were published in 2020 [4]. Consecutive patients with a diagnosis of ICI-OND were prospectively gathered and logged in a database. The study was conducted under Institutional Review Board approval from Memorial Sloan Kettering Cancer Center and complied with the Health Insurance Portability and Accountability Act. All patients provided informed consent. Research adhered to the tenets of the Declaration of Helsinki. Clinical records, including best-corrected Snellen visual acuity, optic nerve description, visual symptoms at diagnosis, other signs of intraocular inflammation, slit lamp and indirect ophthalmoscopy, fundus and optic nerve photography, 24-2 Humphrey visual fields (HVF), Ishihara color vision, retinal nerve fiber layer optical coherence tomography (RNFL OCT), magnetic resonance imaging (MRI), lumbar puncture (LP), and bloodwork results, were reviewed, and relevant data were recorded for each patient eye. If diagnosed by an outside provider before presenting to our institution, those records were reviewed and recorded. Data were compared using the two-tailed Students’ t test with statistical significance set at p value <0.05.
Results
Table 1 summarizes patient demographics, timing of symptoms, treatment summary, and time to follow-up. Of note, patient 15 received treatment for optic nerve inflammation after 126 days of symptoms as this patient had been initially worked up and followed by an outside facility to rule out melanoma-associated retinopathy, before the diagnosis of ICI-induced optic nerve inflammation was made. Patient 17 received treatment for optic nerve inflammation after 52 days of symptoms because he did not report vision symptoms initially. Of note, he was treated for a dermatologic rash with 20 mg prednisone PO × 2 weeks, about 2 weeks after his vision symptoms began. Table 2 summarizes clinical findings, additional testing, and presumed diagnosis. Table 3 outlines patient outcomes and ICI rechallenge. In all patients, the ICI was stopped and steroids commenced except 1 antinuclear antibody seropositive patient (patient 5) who continued ICI and had ONI-OND resolution with oral and subtenon steroids. She then temporarily discontinued ICI for non-ophthalmic reasons and upon re-initiation; she developed rebound optic neuritis with concomitant vitritis, prompting cessation of ICI. She subsequently had resolution and returned to baseline (20/25 and 20/20 vision), at over 3 years follow-up. Four other patients were rechallenged at same ICI dosage, except patient 14 who was rechallenged with a new investigational ICI, without evidence of ICI-OND recurrence. All patients were off steroids at time of rechallenge, except patient 20 who was still tapering steroids.
Table 1.
Patient demographics, timing of symptoms, treatment summary and time to follow-up
| Patient | Age, years | Sex | Alive? (Y/N) | Primary cancer dx | ICI at time of dx | ICI cycles prior to oph dx, n | Time on drug until vision sym, months | Initial ophthalmic treatment* | Time from onset of sym to starting steroid tx, days | Time to follow-up, months |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 58 | M | Y | Renal cell carcinoma | Investigational drug targeting PD-1 and CTLA-4: 750 mg, q3 weeks | 9 | 20 | 80 mg prednisone PO, prednisolone acetate QID OU | 3 | 29.3 |
| 2 | 69 | M | Y | Renal cell carcinoma | Ipilimumab 1 mg/kg + nivolumab 3 mg/kg q3 weeks for 3 cycles, then nivolumab 240 mg q2 weeks | 6 | 4.7 | 80 mg prednisone PO | 1 | 16.2 |
| 3 | 67 | M | Y | Squamous cell carcinoma of the lung | Pembrolizumab 200 mg q3 weeks for 4 cycles, then 400 mg q6 weeks | 5 | 5.5 | 80 mg prednisone PO then 250 mg IV methylprednisone × 3 days | 3 | 11.7 |
| 4 | 70 | M | Y | Nonmetastatic prostate cancer | Pembrolizumab 400 mg q6 weeks | 9 | 13.4 | 60 mg prednisone PO | 4 | 1.2 |
| 5 | 71 | F | Y | Cutaneous melanoma | Nivolumab 480 mg q4 weeks, then 1 dose of ipilimumab 1 mg/kg + nivolumab 3 mg/kg | 1 | 0.2 | 60 mg prednisone PO + subtenon triamcinolone injection (approx. 1 month later) | 5 | 33.8 |
| 6 | 75 | M | Y | Cutaneous melanoma | Nivolumab 480 mg q4 weeks | 2 | 0.1 | 100 mg prednisone PO | 7 | 53.6 |
| 7 | 67 | F | Y | Cutaneous melanoma | Ipilimumab 1 mg/kg + nivolumab 3 mg/kg q3 weeks | 2 | 1.6 | 1 mg/kg PO prednisone | 7 | 28.7 |
| 8 | 43 | M | N | Renal cell carcinoma | Ipilimumab 1 mg/kg, nivolumab 3 mg/kg q3 weeks | 3 | 2.1 | Prednisone 80 mg PO, prednisolone acetate QID, diflurprednate QID | 7 | 29.4 |
| 9 | 39 | F | Y | Invasive ductal carcinoma | Pembrolizumab 400 mg q6 weeks | 5 | 7.5 | IV methylprednsone 1,000 mg q24 h × 3 days + prednisolone acetate QID | 7 | 4.0 |
| 10 | 61 | F | Y | Cutaneous melanoma | Ipilimumab 3 mg/kg + nivolumab 1 mg/kg q3 weeks | 3 | 1.7 | Prednisone 80 mg PO, prednisolone acetate q1h | 10 | 91.7 |
| 11 | 75 | M | Y | Lung adenocarcinoma | Pembrolizumab 200 mg q4 weeks | 7 | 5.0 | 60 mg prednisone PO | 14 | 4.1 |
| 12 | 71 | F | N | Lung adenocarcinoma | Investigational drug targeting PD-1 and CTLA-4: 3 mg/kg, q2 weeks | 6 | 2.6 | 60 mg prednisone PO | 7 | 1.7 |
| 13 | 82 | M | Y | Cutaneous melanoma | ipilimumab 1 mg/kg + nivolumab 3 mg/kg q3 weeks | 2 | 4.7 | 60 mg prednisone PO | 0 | 3.0 |
| 14 | 68 | M | Y | Cutaneous melanoma | Ipilimumab 3 mg/kg + nivolumab 1 mg/kg q3 weeks | 4 | 3.8 | IV methylprednisone × 3 days, prednisone 1 mg/kg PO | 2 | 59.5 |
| 15 | 58 | F | Y | Cutaneous melanoma | Ipilimumab 3 mg/kg + nivolumab 1 mg/kg q3 weeks, followed by nivolumab | 52 | 38.6 | IV dexamethasone 3 g, IV immunoglobulin × 1, rituximab × 3, plasma exchange × 5, prednisone 80 mg PO | 126 | 12.1 |
| 16 | 59 | F | N | SCLC | Nivolumab 240 mg q3 weeks | 2 | 0.7 | Prednisone 60 mg PO | 0 | 7.0 |
| 17 | 71 | M | N | NSCLC | Pembrolizumab 200 mg q4 weeks | 3 | 4.5 | IV methylprednisone 1 g × 5 days, prednisone 80 mg PO | 52 | 6.5 |
| 18 | 54 | F | N | NSCLC | Ipilimumab 1 mg/kg q6 weeks and nivolumab 240 mg q3 weeks | 4 | 1.8 | Prednisone 80 mg PO | 4 | 2.5 |
| 19 | 73 | F | Y | Renal cell carcinoma | Atezolizumab 1,200 mg q3 weeks | 95 | 66.4 | Prednisone 80 mg PO | 31 | 1.6 |
| 20 | 66 | F | N | Invasive ductal carcinoma | Ipilimumab 1 mg/kg + nivolumab 240 mg q6 weeks and nivolumab 240 mg q3 weeks | 11 | 5.1 | Prednisone 60 mg PO | 0 | 4.7 |
| Statistical analysis | | | | | | Median = 4.5 cycles | Median = 4.5 months | | Median = 6 days | Median = 9.4 months |
Dx, diagnosis; ICI, immune checkpoint inhibitor; oph, ophthalmic; sym, symptoms; tx, treatment; PD-1, programmed cell death protein 1; CTLA-4; cytotoxic T-lymphocyte antigen 4; PO, oral; QID, four times a day; OU, both eyes; IV, intravenous; q1h, every hour; SCLC, small cell lung carcinoma; NSCLC, non-small cell lung carcinoma.
aPatient 15 received treatment for optic nerve inflammation after 126 days of symptoms as this patient had been initially worked up and followed by an outside facility to rule out melanoma-associated retinopathy, before the diagnosis of ICI-induced optic nerve inflammation was made. Patient 17 received treatment for optic nerve inflammation after 52 days of symptoms because he did not report vision symptoms initially. Of note, he was treated for a dermatologic rash with 20 mg prednisone PO × 2 weeks, about 2 weeks after his vision symptoms began.
Table 2.
Clincial findings, additional testing and presumed diagnosis
| Patient | Optic nerve description | Symptoms at diagnosis | Other intraocular signs of inflammation | Color vision deficit at presentation (Y/N) | VF abnormality at diagnosis | RNFL OCT flagged as abnormal on TSNIT map (Y/N) | MRI abnormality at diagnosisa | Additional systemic work-upb | Known concurrent autoimmune disease (Y/N) | Presumed diagnosis |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Superonasal optic nerve edema OS | Blurred vision, redness | Choroiditis, uveitis, vitritis OU | NR | OS: cecocentral | N | No abnormality | (−) RPR, lyme, quant TB | N | ICI-panuveitis with ICI-optic nerve edema |
| 2 | Optic nerve edema OS>OD | Gray cloud in vision | None | Y | OD: normal | Y | No abnormality | (−) ESR, ANA, ANCA, ACE, RPR, lyme | N | ICI-optic nerve edema |
| OS: cecocentral | ||||||||||
| 3 | Hemorrhagic optic nerve edema OU | Blurred vision | choroiditis OU | Y | OD: normal | Y | Edema within left intraorbital nerve | (−) ESR, CRP, ANA, ANCA, ACE, RPR, lyme | N | ICI-optic neuritis and ICI-posterior uveitis |
| OS: inferior altitudinal | ||||||||||
| 4 | Normal OU | Blurred vision | None | Y | OD: central defect | N | Marked left intraorbital optic nerve and perineural enhancement, possible right perineural enhancement | (−) ESR, CRP | N | ICI-optic neuritis |
| OS: superior altitudinal and inferotemporal | ||||||||||
| 5 | sectoral optic nerve edema OS | Blurred vision, pain, redness | Uveitis and vitritis OU | NR | NR | Y | No abnormality | (−) Vitreous biopsy, RF, RPR, lyme, HSV1/2 | N | ICI-panuveitis with ICI-optic nerve edema |
| (+) ANA (titer 1.160, speckled pattern) | ||||||||||
| 6 | Optic nerve pallor OD (outside dx) | Blurred vision, photosensitivity | None | Y | OD: central defect | Y | NR | (−) B12, folate | N | ICI-optic nerve edema |
| 7 | Hemorrhagic optic nerve edema OS>OD | Blurred vision | Vitritis OS | Y | Inferior altitudinal OU | Y | Minimal subtle enhancement around the optic disc OS | (−) ESR, CRP, ANA, ANCA, ACE, RPR, lyme, quant TB, NMO, MRV, MRA | N | ICI-posterior uveitis with ICI-optic nerve edema |
| 8 | Mild optic nerve fullness OU | Blurred vision, pain, redness | Panuveitis OU | N | non-specific OU | Y | NR | None | N | ICI-panuveitis with ICI-optic nerve edema |
| 9 | Optic nerve edema OU | Blurred vision, pain | Choroiditis OU | Y | OD: superior defect | Y | Optic perineuritis and scleritis OU | (−) RPR, lyme, quant TB, HIV, NMO and MOG Ab | N | ICI-perineuritis and ICI-posterior uveitis |
| OS: non-specific | ||||||||||
| 10 | Mild nasal optic nerve edema OU | Blurred vision, floaters | Anterior uveitis OU | Y | OD: diffuse defect | Y | Subtle optic nerve and optic nerve sheath changes | None | Y: psoriatic arthritis | ICI-optic neuritis and ICI-anterior uveitis |
| OS: No abnormality | ||||||||||
| 11 | Mild temporal optic nerve edema OU | Grayed out vision, peripheral scotoma | None | N | Non-specific defects OU | N | Subtle optic nerve abnormality vs. motion artifact | (−) Carotid doppler, head and neck CTA | N | ICI-optic neuritis |
| 12 | Superior optic nerve edema OU | Blurred vision, gray spot | Anterior uveitis, choroiditis OU | Y | Cecocentral defect OU | Y | Bilateral optic nerve edema, no abnormal enhancement | (−) ANA, ANCA, RF, ACE, RPR, lyme, quant TB | N | ICI-optic nerve edema and ICI-panuveitis |
| 13 | Optic nerve edema OU | Blurred vision | None | N | OD: superior defect | Y | Enhancement of optic nerve head and distention of optic nerve sheath | (−) ESR, CRP, RF, ANA, ANCA, ACE, RPR, lyme, quant TB, bartonella | N | ICI-optic neuritis |
| OS: central defect | ||||||||||
| 14 | Superonasal optic nerve edema OD | Peripheral scotoma | None | N | OD: inferior altitudinal | Y | No abnormality | none | N | ICI-optic nerve edema |
| 15 | Mild optic nerve edema OU | Blurred vision, floaters, halos | None | Y | OD: non-specific | NR | No abnormality | none | Y: ICI-induced sarcoidosis (inactive) | ICI-optic nerve edema |
| OS: central | ||||||||||
| 16 | Hemorrhagic optic nerve edema OS | Blurred vision, floaters | None | NR | OS: inferior altitudinal | Y | No abnormality | none | N | ICI-optic nerve edema |
| 17 | Optic nerve edema OU | Blurred vision | None | Y | Diffuse OU | Y | Papilledema, bilateral, enhancement surrounding intraorbital optic nerve sheaths | none | N | ICI-optic neuritis |
| 18 | Optic nerve edema OU | Blurred vision | Choroiditis OU | Y | Incomplete bitemporal defect OU | Y | Mild enlargement of left greater than right distal intraorbital optic nerve | none | N | ICI-optic neuritis and ICI-posterior uveitis |
| 19 | Optic nerve edema OU | Floaters | Anterior uveitis and vitritis OU | N | No abnormality OU | Y | No abnormality | none | N | ICI-panuveitis with ICI-optic nerve edmea |
| 20 | Optic nerve fullness OU | Blurred vision | None | Y | No abnormality OU | Y | No abnormality | none | N | ICI-optic nerve edema |
| Statistical analysis | 70% of patients had bilateral involvement | 100% symptomatic | 50% had concomitant intraocular inflammation | 59% of eyes abnormal (none with known congenital defect) | 79% of eyes had defects (18% altitudinal) (15% non-specific) (12% central) (12% cecocentral) | Of 35 eyes assessed, 80% had abnormal RNFL (86% of these thicker than avg) | Of 18 patients who received MRI, 56% were abnormal | | | |
OD, right eye; OS, left eye; OU, both eyes; VF, visual field; RNFL OCT, Retinal nerve fiber layer optical coherence tomography; MRI, magnetic resonance imaging (3 Tesla MRI of brain and or orbits with and without contrast); NR, not recorded; RPR, rapid plasma reagin; quant TB, quantiferon-TB Gold; ESR, erythrocyte sedimentation rate; ANA, antinuclear antibody; ANCA, antineutrophil cytoplastic antibodies; ACE, angiotensin converting enzyme; CRP, C-reactive protein; RF, rheumatoid factor; HSV1/2, Herpes simplex virus 1 and 2; HIV, human immunodeficiency virus; NMO, neuromyelitis optica; MOG Ab, myelin oligodendrocyte glycoprotein antibody; CTA, computed tomography angiography; MRA, magnetic resonance angiography; MRV, magnetic resonance venography.
a18 patients underwent MRI brain and/or orbits at presentation and 56% had abnormal findings, one of whom had no visible optic nerve abnormality, but poor HVFs prompted MRI imaging which revealed intraorbital and perineural enhancement. The remaining 8 patients presented with ophthalmoscopically evident optic nerve abnormality without retrobulbar abnormalities on MRI imaging. Of the patients with abnormal MRI findings, 60% involved the optic nerve head, 30% involved the intraorbital nerve, and 30% involved the optic nerve sheath (some patients had more than one area of abnormality on MRI).
bAdditional bloodwork and or imaging (including carotid doppler, computed tomography angiography, magnetic resonance venography, and angiography) and or vitreous biopsy were performed in 10 patients, and only 1 patient had an initial positive ANA finding (patient 5). This patient did not have any systemic signs/symptoms of lupus but was sent to be worked up with a rheumatologist and has not had further work-up to date, despite repeated recommendations. Two other patients had known (but controlled or inactive) concomitant autoimmune disease: patient 10 had psoriatic arthritis managed with methotrexate and secukinumab, and patient 15 developed biopsy proven ICI-induced sarcoidosis of the skin and lungs 3 years prior and was inactive at the time of ocular diagnosis.
Table 3.
Patient outcomes and ICI rechallenge
| Patient | Combination vs. monotherapy ICI at time of optic neuritis treatment | Vision better than or equal to 20/40 in at least one affected eye after treatment? (Y/N)a | Normal color vision after treatment? (Y/N)b | RNFL OCT flagged as abnormal after treatment? (Y/N) | Rechallenge with ICI? | Was patient on steroids at time of rechallenge? | Time to follow-up since ICI rechallenge, months |
|---|---|---|---|---|---|---|---|
| Was there recurrence of optic neuritis with rechallenge?c | |||||||
| 1 | Combo | Y | NR | N | N | -- | -- |
| 2 | Mono | Y | Y | Y | Y (nivolumab 240 mg q 2 weeks, then 480 mg monthly) | N;N | 4.0 |
| 3 | Mono | N | NR | Y | N | -- | -- |
| 4 | Mono | Y | Y | N | N | -- | -- |
| 5 | Combo | Y | Y | Y | Y (ipilimumab 1 mg/kg + nivolumab 3 mg/kg) | N;Y | 12.9 |
| 6 | Mono | Y | Y | Y | N | -- | -- |
| 7 | Combo | Y | N | Y | N | -- | -- |
| 8 | Combo | Y | Y (normal at baseline) | Y | N | | -- |
| 9 | Mono | Y | Y | N | N | -- | -- |
| 10 | Combo | Y | Y | Y | N | -- | -- |
| 11 | Mono | Y | Y (normal at baseline) | Y | Y (pembrolizumab 200 mg q 3 weeks) | N;N | 2.7 |
| 12 | Combo | Y | N | Y | N | -- | -- |
| 13 | Combo | Y | Y (normal at baseline) | Y | N | -- | -- |
| 14 | Combo | N | Y (normal at baseline) | Y | Y (PD-1 investigational drug q 3 weeks) | N;N | 17 |
| 15 | Mono | Y | N | Y (but no baseline) | N | -- | -- |
| 16 | Mono | N | NR | Y | N | -- | -- |
| 17 | Mono | Y | N | Y | N | -- | -- |
| 18 | Combo | Y | NR | Y | N | -- | -- |
| 19 | Mono | Y | Y (normal at baseline) | Y | N | -- | -- |
| 20 | Combo | Y | Y | Y | Y (ipilimumab 1 mg/kg + nivolumab 240 mg q 6 weeks and nivolumab 240 mg q 3 weeks) | Y;N | 2.3 |
| Statistical analysis | | 89% had 20/40 vision in at least one eye | Of eyes with initial dyschromatopsia, 53% regained normal color vision | Of eyes with thickened RNFL, 44% had less thickening (but still abnormal) after tx | | 6 involved eyes were rechallenged, no recurrence in 88% | |
| Median time off ICI before rechallenge was 2.6 months |
ICI, immune checkpoint inhibitor; RNFL OCT, retinal nerve fiber layer optical coherence tomography.
aThe 3 patients whom vision was not better than 20/40 did receive prompt steroid treatment after onset of visual symptoms (day 3, day 2, and day 0) and 2 of the 3 had been on single-agent ICI at the time of optic nerve inflammation.
bOnly 7% (2 eyes) were thickened at presentation and had more thickening of RNFL after steroid treatment (both of which had incorrectly mapped RNFL scans at baseline, thereby making comparison difficult).
cThe one eye (patient 5) that developed a recurrence of optic nerve inflammation on ICI was initially treated with oral steroids while continuing on ICI therapy. After resolution of optic nerve inflammation, ICI was discontinued (for non-ophthalmic reasons), then resumed (ipilimumab + nivolumab) 3 months later, and the patient developed recurrent vitritis in both eyes and optic nerve edema in one eye. ICI was discontinued once again and the optic nerve inflammation was treated with a course of oral steroids – she has not resumed ICI therapy since. Of note, this was the same patient with positive ANA finding.
Most eyes achieved improvement in vision after completion of steroid treatment: 31 (84%) eyes had improved vision, 4 (11%) eyes maintained stable vision, and only 2 eyes had worsening of vision (no more than 1 Snellen line) after steroids. Eyes with initial vision of 20/40 or worse improved 3.7 (mean) lines of Snellen acuity after steroid treatment and eyes with initial vision of 20/30 or better improved 0.71 (mean) lines of Snellen acuity after steroid treatment (p = 0.0002).
Discussion
Recent classification systems have been introduced to help stratify ICI-induced optic nerve disorders [5]. Our cohort of patients with various ICI-OND reveals that the majority presented with bilateral (70%), painless (85%) visual decline and color vision abnormality (59%); after a median of 4.5 cycles of ICI therapy. ICI toxicity involving other organs may occur over a wide time range, namely, within days, weeks or months after treatment, even after ICI cessation (median onset of toxicity is 2–16 weeks). For ICI-OND, the onset can be longer, and therefore, this diagnosis should be considered in cases of prolonged ICI use [6]. Only 3 patients experienced eye pain, and all had concomitant intraocular inflammation– suggesting the optic nerve inflammation may not be a consistent source of pain [7]. Most patients presented with visible optic nerve edema – except patient 4, where MRI aided in the diagnosis of retrobulbar inflammation (Fig. 1a, b) and patient 6 who had been diagnosed and treated with oral steroids before presentation to our clinic and, therefore, had the appearance of optic nerve pallor upon transfer.
Fig. 1.
a Fundus images of patient 4 with normal appearing optic nerves. b An MRI orbits of the same patient revealing marked abnormal long segment enhancement within and along the intraorbital course of the left optic nerve, outlined with red arrow. c A thickened RNFL scan of patient 10. d An MRI orbits of the same patient revealing subtle central T2 hyperintense changes of the bilateral intraorbital optic nerves and optic nerve sheath, outlined with red arrows.
Ancillary testing such as HVFs and MRI can aid in diagnosis, particularly with subtle optic nerve changes, but cannot be relied upon solely for diagnosis [4]. This cohort reinforces these conclusions: HVF defects were highly variable and MRI findings aided in diagnosis in 56% patients. This is in contrast to classic optic neuritis where 66% of patients have diffuse loss [8] and up to 55% of patients with ischemic optic neuropathy presenting with altitudinal loss [9]. To this point, we demonstrate RNFL OCT is useful in demonstrating nerve fiber layer edema. This can be of particular use when optic nerve margins are subtly edematous (Fig. 1c, d). Other isolated reports in the literature describe patients with similar features [10–13].
Concurrent autoimmune disease may influence disease course. As seen in Table 3, if concomitant autoimmune disease is well managed, vision outcomes may be favorable (patient 10). In contrast, patient 5 had unaddressed autoimmune disease and was the only patient to develop recurrent ICI-OND following drug rechallenge, suggesting concurrent uncontrolled autoimmune disease may exacerbate ICI toxicity and may make ICI rechallenge difficult. Two of the 5 patients who were rechallenged, were started on combination ICI therapy. One of those patients was on steroids at time of rechallenge and did not recur. The other (patient 5) was not on steroids during rechallenge and recurred while on combination ICI. This may indicate that in patients who are rechallenged with combination ICI and have concomitant uncontrolled autoimmune disease may be more prone to developing ICI-OND recurrence.
Systemic steroids and ICI cessation can improve and maintain vision (84%; 89% of eyes with 20/40 vision or better) and possibly color vision (53%) at follow-up (mean 20 months; range 1.2–91.7 months) – even after steroid completion. It is not surprising that vision improved in more patients compared to color vision, particularly because vision measurements rely on high contrast charts (digital screen) and may not fully reflect deficits with low contrast viewing – an issue more easily reflected in color vision testing. Interestingly, there may be a limit to visual recovery despite prompt steroid initiation, as evidenced by 3 patients who received prompt steroid treatment (within 3 days of symptom onset) and did not achieve vision better than 20/40. Lastly, we can confirm that although a smaller sub-cohort (8 eyes), most eyes (88%) had no detectable recurrence (mean 8 months follow-up) with ICI rechallenge after receiving anywhere from 4 to 11 cycles of ICI, only one of which was on oral steroids at time of rechallenge. Since this cohort expands on a smaller (multicenter) cohort [4], there is some variation in the administration route and dosage of steroids for the treatment of ICI-OND. Additionally, some of these patients were seen as early as 2017, where there was little consensus in the literature regarding treatment of ICI-OND, hence the variability in steroid route and administration.
ICI is more commonly known to cause other systemic inflammation such as colitis (40%), followed by hepatitis at 30%; and the treatment for these toxicities starts with corticosteroids and a slow taper. A retrospective study involving 80 melanoma patients with various immune-related toxicities were rechallenged with ICI monotherapy and recurrent toxicity was reported in 18–21% of patients [14]. Consistent with this, our cohort identified recurrence of ICI-OND in 1 of 5 patients after ICI rechallenge. Longer follow-up (and a larger sample size) will help paint a clearer picture about ICI rechallenge. We acknowledge that as a tertiary referral center, there is a degree of sample bias and may not generalize to different socioeconomic groups or earlier stage of disease. In that vein, it is important to address how this translates to care across the globe, particularly in third world countries, where there is a lack of resources. For example, high-resolution MRI imaging may not be readily available in underserved communities. However, our findings suggest that the results of these tests did not always aide in a diagnosis (MRI findings were diagnostic in 56% patients). Reassuringly, our findings suggest that treatment with high dose oral steroids (which is readily available globally) is an effective way of treating this spectrum of disorders. Even in the most underserved communities with no ophthalmic imaging available, the presence of uveitis and optic nerve inflammation can be diagnosed with a slit lamp examination, and therefore, treatment can be initiated with little to lose and much to gain for these patients.
In the majority of cases, ICI-OND occurs bilaterally with painless vision decline and dyschromatopsia. Cessation of drug along with initiation of systemic steroids can improve vision to 20/40 or better in at least one involved eye and although a smaller subset of patients, we found that ICI can be rechallenged after completion of steroids, without documented recurrence of ICI-OND in 80% of patients.
Statement of Ethics
The study was conducted under Institutional Review Board approval, ID: 20-126, from Memorial Sloan Kettering Cancer Center and complied with the Health Insurance Portability and Accountability Act. All patients provided written informed consent to participate. Written informed consent was obtained from the patient for publication of the details of their medical case and any accompanying images. Research adhered to the tenets of the Declaration of Helsinki.
Conflict of Interest Statement
The authors have no conflicts of interest to declare.
Funding Sources
The study was supported in part by funding from the Fund for Ophthalmic Knowledge and the New York Community Trust. The funding agencies had no role in the design or conduct of this research. This research was funded in part through the NIH/NCI Cancer Center Support Grant (P30 CA008748).
Author Contributions
J.C.: design of work, acquisition, analysis and interpretation of data, drafting the work, final approval, and agreement to be accountable for all aspects of the work. R.F.S., D.H.A., and J.H.F.: conception/design of work, critical review, final approval, and agreement to be accountable for all aspects of the work.
Funding Statement
The study was supported in part by funding from the Fund for Ophthalmic Knowledge and the New York Community Trust. The funding agencies had no role in the design or conduct of this research. This research was funded in part through the NIH/NCI Cancer Center Support Grant (P30 CA008748).
Data Availability Statement
All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.
References
- 1. Dhasmana A, Dhasmana S, Haque S, Cobos E, Yallapu MM, Chauhan SC. Next-generation immune checkpoint inhibitors as promising functional molecules in cancer therapeutics. Cancer Metastasis Rev. 2023;42(3):597–600. [DOI] [PubMed] [Google Scholar]
- 2. Liu C, Francis J, Pulido J, Francis J, Abramson D. Ocular side effects of systemically administered chemotherapy; 2024. Available from: https://www.uptodate.com/contents/ocular-side-effects-of-systemically-administered-chemotherapy (accessed July 13, 2024). [Google Scholar]
- 3. Dalvin L, Shields C, Orloff M, Sato T, Shields JA. Checkpoint inhibitor immune therapy, systemic indications and ophthalmic side effects. Retina. 2018;38(6):1063–78. [DOI] [PubMed] [Google Scholar]
- 4. Francis J, Jaben K, Santomasso B, Canestraro J, Abramson DH, Chapman PB, et al. Immune checkpoint inhibitor-associated optic neuritis. Ophthalmology. 2020;127(11):1585–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Chang EL, Liu R, Keyhanian K, Huynh K, Berkenstock M, Bhatti MT, et al. Consensus disease definitions for ophthalmic immune-related adverse events of immune checkpoint inhibitors. J Immunother Cancer. 2025;13(4):e011049. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Gumusay O, Callan J, Rugo HS. Immunotherapy toxicity: identification and management. Breast Cancer Res Treat. 2022;192(1):1–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Optic Neuritis Study Group . The clinical profile of acute optic neuritis: experience of the optic neuritis treatment trial. Arch Ophthalmol. 1991;109:1673–8. [DOI] [PubMed] [Google Scholar]
- 8. Keltner J, Johnson C, Cello K, Dontchev M, Gal RL, Beck RW, et al. Optic neuritis study group. Visual field profile of optic neuritis: a final follow-up report from the optic neuritis treatment trial from baseline through 15 years. Arch Ophthalmol. 2010;128(3):330–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Rizzo J, Lessell S. Optic neuritis and ischemic optic neuropathy. Overlapping clinical profiles. Arch Ophthalmol. 1991;109(12):1668–72. [DOI] [PubMed] [Google Scholar]
- 10. Cuzzubbo S, Philippakis E, Baroudjian B, Ursu R, Lebbé C, Coutourier A, et al. Optic neuropathy related to immune checkpoint inhibitors: a case series. Neuro Oncol. 2023;25(Suppl ment_2):49. [Google Scholar]
- 11. Daetwyler E, Zippelius A, Meyer P, Läubli H. Pembrolizumab-induced optic neuropathy – a case report. Front Immunol. 2023;14:1171981. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Nakajima I, Nakaizumi T, Tsuji H. Optic neuritis as an immune-related adverse event of pembrolizumab. Neurol Clin Neurosci. 2024;12(5):325–6. [Google Scholar]
- 13. Zhang J, Wang L, Zhang L, Liu H, Yang B, Jiang L. Bilateral retrobulbar optic neuritis combined with optic perineuritis associated with atezolizumab treatment for small cell lung carcinoma. Eur J Ophthalmol. 2024;34(3):NP29–33. [DOI] [PubMed] [Google Scholar]
- 14. Pollack MH, Betof A, Dearden H, Rapazzo K, Valentine I, Brohl AS, et al. Safety of resuming anti-PD-1 in patients with immune-related adverse events (irAEs) during combined anti-CTLA-4 and anti-PD1 in metastatic melanoma. Ann Oncol. 2018;29(1):250–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.