Back Back
FAQs: ADC Ocular Toxicity
Answering Your Questions: Ocular Toxicity With Antibody–Drug Conjugates

Released: January 25, 2021

Expiration: January 24, 2022

Asim V. Farooq
Asim V. Farooq, MD
Sagar Lonial
Sagar Lonial, MD, FACP
Anthony J. Perissinotti
Anthony J. Perissinotti, PharmD, BCOP
Activity Information

Activity

Progress
1
Course Completed

Ophthalmologists and optometrists have a critical role in the multidisciplinary management of patients with cancer who are receiving antibody–drug conjugates (ADCs) that are associated with ocular toxicity. Collaboration between eyecare professionals and the oncology care team is essential to ensuring optimal outcomes from therapy by minimizing and preventing any potential corneal toxicity that is an off-target effect of some of the cytotoxic payloads used with ADCs. In particular, the cytotoxic payload monomethyl auristatin F that is a component of the ADC belantamab mafodotin, which was recently approved for patients with relapsed/refractory multiple myeloma as well as other investigational ADCs, has been associated with a relatively high incidence of ocular adverse events that require appropriate monitoring and management. In a recent live Webinar with my oncology colleagues, Sagar Lonial, MD, and Anthony Perissinotti, PharmD, BCOP, we described the mechanism of this ocular toxicity and strategies to monitor for the development of it and how to manage symptoms to prevent worsening and enable patients to continue to receive their therapy.

Here, my colleagues and I address several questions we received from our audience during the live event.

Which ADCs cause corneal toxicity?

Asim V. Farooq, MD: Theoretically, based on the proposed pathophysiology, any ADC has the potential to cause corneal epithelial toxicity. In terms of what has been published in the literature, in addition to belantamab mafodotin, trastuzumab emtansine and investigational agents such as depatuxizumab mafodotin (ABT-414), denintuzumab mafodotin (SGN-CD19A), vorsetuzumab mafodotin (SGN-75), and mirvetuximab soravtansine (IMGN853) have been reported to cause corneal toxicity. Belantamab mafodotin is the only approved ADC with a Risk Evaluation and Mitigation Strategy for ocular toxicity, but I’ve seen patients on clinical trials for a number of other ADCs with virtually identical changes in the cornea. 

The ocular toxicity may be related to multiple factors that can include: (1) whether or not the corneal epithelium expresses the target of interest, (2) the nature of the linker (eg, noncleavable vs cleavable linker), and (3) the nature of the cytotoxic payload. In theory, there are multiple pathways by which an ADC could potentially cause corneal epithelial toxicity. As we covered in the Webinar, there is on-target toxicity and off-target toxicity. Off-target toxicity can be Fc receptor mediated toxicity, pinocytosis, or bystander toxicity. It really depends on these factors, but certainly ocular toxicity is possible with any ADC.

How long does it typically take for patients to come back to grade 1 or baseline with keratopathy-related events? Is there an average time or average number of dose holds?

Asim V. Farooq, MD: It depends on the degree of toxicity initially. For example, a grade 2 event may revert back to grade 1 relatively quickly compared with grade 3 or 4 toxicities, which will usually take longer. From patients whom I’ve seen who were being treated with belantamab mafodotin and the data that we’ve observed, patients will typically present with microcyst-like epithelial changes (MECs) between Day 30 and Day 40. Improving to grade 1 can vary in the range of 3-6 weeks. Again, it depends on the grade of toxicity at presentation. Among patients who have undergone dose delays, in my anecdotal experience, the corneal findings are often resolved by the 6-week time point.

Sagar Lonial, MD: Some of this also depends on differentiating between changes in visual acuity and changes in keratopathy. Keratopathy is something that starts on the periphery and tends to work its way centrally, which is why the visual acuity changes tend to occur later on. In the same fashion, reversal of visual acuity changes tend to start in the central area and then work peripherally. The average time for improvement in changes in visual acuity is approximately 22 days, whereas the average time to recovery of keratopathy is probably closer to 60-70 days. This explains the disconnect between an exam finding of keratopathy and clinical symptoms. I believe it is important to highlight with our patients who are being treated with belantamab mafodotin that clinical symptoms that have significant impact on visual acuity only occurred 18% of the time, whereas keratopathy occurred 70% of the time, and the majority of those were asymptomatic or symptomatic with dry eyes or itchy eyes. Therefore, it is very important to distinguish the exam finding of keratopathy between physical changes that they may experience in their vision, which is a much more rare occurrence.

Asim V. Farooq, MD: Absolutely, and that is something that we’ve seen in the data and I’ve seen a number of times now in clinical practice, where I will tell a patient, “You have these microcyst-like changes,” but their vision is really fine. Thus, it is very important to differentiate what is happening and to monitor them very closely, as there is this disconnect between the presence of these lesions and the impact on visual acuity that we believe is related to the location as well as concentration of the lesions. If they are in the periphery, they are much less likely to cause any symptoms.

If blurry vision does not improve or worsens after holding the treatment for 6 weeks, would steroid drops or other management be recommended?

Asim V. Farooq, MD: Steroid drops were assessed in the ocular substudy of the DREAMM-2 trial with belantamab mafodotin, and as a prophylaxis, they were not found to be helpful in preventing the presence of MECs. I would say that, based on our understanding of the mechanism of action and based on what we’ve observed, steroids are not likely, in my opinion, to lead to a reversal of the MECs. And I would say that—in the vast majority of patients that we’ve seen—the visual acuity and the keratopathy does improve over a period of time (approximately 6 weeks), and if it hasn’t completely resolved, then it may mean that the drug needs to be held longer. I haven’t seen any data nor observed any patients who experienced worsening visual acuity beyond this time point. Remember, the proposed mechanism of action and pathophysiology is that this is a dose-dependent toxicity related to the Ctau. Preservative-free artificial tears should be recommended. For some patients who have whorl-like staining that may be indicative of limbal stem cell dysfunction, steroids can be considered and I have recommended this. But I would not necessarily initiate steroids for the presence of MECs.

Sagar Lonial, MD: As Dr. Farooq pointed out, there was a randomization in DREAMM-2 that randomized patients to steroids vs no steroids as a way to prevent the development of keratopathy and that randomization demonstrated no benefit to steroid eye drops. So I think, unless an ophthalmologist with experience like Dr. Farooq is recommending it be done, it should not be done empirically.

How effective are the eyedrops in preventing this toxicity?

Asim V. Farooq, MD: At present, we do not have any evidence that an eye drop prevents MECs that occur with belantamab mafodotin. Prophylactic topical steroids were studied in DREAMM-2 and were not found to prevent these changes. Of note, a history of dry eye was correlated with a higher risk of grade ≥ 2 corneal findings; therefore, lubrication with preservative-free artificial tears should be recommended, especially among patients with dry eye. It is not known at present whether aggressive lubrication or other measures to treat dry eye are effective for prevention.

What if the drop in best-corrected visual acuity (BCVA) is not due to the cornea?

Asim V. Farooq, MD: The only changes that were found in DREAMM-2 that were felt to be drug related were corneal. Patients with multiple myeloma may have a number of other ocular conditions that may affect their vision (unrelated to belantamab mafodotin). These may include cataract, advanced glaucoma, or macular degeneration. When an eye care provider is seeing a patient who is receiving belantamab mafodotin and a drop in vision cannot be explained by corneal changes (that are drug related), other causes should be investigated. Nondrug-related causes of decreased vision would not generally lead to a dose hold or dose modification.

Is it possible to have mild keratopathy WITH a 3-line change in that same eye? How would that patient be graded?

Asim V. Farooq, MD: It is theoretically possible, although this is not the typical scenario. I should add that, in terms of the Keratopathy Visual Acuity (KVA) scale for dose hold/modification, the decrease in vision should be drug related. If there is a discordance between the keratopathy and the vision change, then the WORSE finding between the two should be used for grading.

How would you grade the patient if their baseline bilateral BCVA is 20/200.

Asim V. Farooq, MD: This is a good question. The KVA scale was created with the understanding that some patients will have a BCVA of 20/20, whereas others will have a lower BCVA due to various ocular conditions (eg, cataract, macular degeneration). The grades were created to account for a decrease in visual acuity caused by belantamab mafodotin, while taking into consideration the effect on a patient’s activities of daily living. For example, a patient with baseline BCVA of 20/20 who experiences a 2-line decrease in vision may still be able to drive, whereas a patient with a baseline BCVA of 20/50 who experiences a 2-line decrease in vision may no longer be able to drive.

Using the KVA scale as a guide, the patient, hematologist, and eye care provider must decide together regarding the decision to continue or delay treatment. Regarding the specific question of a patient with a baseline BCVA of 20/200, technically, according to the KVA scale, any decline in BCVA below 20/200 would qualify as grade 4; the visual acuity would need to return to baseline (ie, 20/200) and the keratopathy would need to be resolved before further dosing (at a reduced dose).

When you see a 3-line drop in a patient’s visual acuity on Snellen, what ocular exam do you perform next to get the BCVA?

Asim V. Farooq, MD: A refraction is performed by an eye care provider (or qualified technician in an eye care provider’s office).

Are there acuity changes or corneal changes that would jeopardize the patients’ driving safety or cause you to advise them not to drive?

Asim V. Farooq, MD: This really depends on the patient’s baseline visual acuity and the change in visual acuity that occurs during treatment. For example, a patient with a baseline visual acuity of 20/20 may be able to tolerate corneal changes and an accompanying decline in visual acuity (eg, to 20/40), whereas a patient with a baseline visual acuity of 20/50 may not be able to drive if visual acuity declines. Patients should be advised to use caution when driving or using heavy machinery during treatment. Visual acuity measurements obtained at the eye care provider’s office will be informative, and the visual acuity requirements for driving (which in the United States vary from state to state) should be reviewed. A patient should also be advised not to drive if he or she feels uncomfortable doing so.

On average, what is time latency prior to developing ocular toxicity (after which dose) and average time frame of recovery before reinitiation of drug?  

Asim V. Farooq, MD: The median time to occurrence of MECs in DREAMM-2 was 37 days. I am unsure as to when the drug was reinitiated on average. Although the first MEC event lasted a median of 86.5 days and the visual acuity change lasted a median of 33 days, per the KVA scale, patients could be restarted on the drug after improving to grade 1.

Were patients in the DREAMM-2 trial prohibited from wearing contacts, or is this an additional safety measure compared with patients in the trial?

Asim V. Farooq, MD: Patients in DREAMM-2 did not wear contacts, partly due to safety concerns, but also due to consideration of having a standardized protocol, as is typical for clinical trials.

Is there any correlation between ocular toxicity and tumor response?

Asim V. Farooq, MD: This is a good question. It is possible that the manufacturer of belantamab mafodotin has access to this information, but I am unaware of the answer to this question. I also have not seen this information for other ADCs. I suspect that, if there is a correlation, it may be weaker for off-target toxicity (ie, when the target antigen is not expressed in the cornea, such as with belantamab mafodotin) compared with on-target toxicity.

Any absolute ophthalmologic contraindication to start therapy with belantamab mafodotin?

Asim V. Farooq, MD: In DREAMM-2, patients with ocular surface disease beyond mild dry eye were excluded. The effect of these corneal findings on patients with more severe ocular surface disease is currently unknown. However, to my knowledge, there are no absolute contraindications to start therapy. Given that patients being started on belantamab mafodotin have failed therapy with multiple agents for multiple myeloma, the risks/benefits should be discussed with the patient, including the ocular adverse events.

Any experience with cold packs during infusion lessening ocular effects?

Asim V. Farooq, MD: I do not have experience with this. I do know that this has been proposed (and is being used in some clinical trials with other ADCs) as a means of causing local vasoconstriction in an attempt to limit the amount of drug that reaches the cornea via the limbal blood vessels. However, given that the drug circulates in the bloodstream (beyond just at the time of infusion), I feel that this may not be effective in preventing ocular adverse events seen with ADCs.

Are there any other tissues in the body that are similar to corneal tissue and thus also susceptible, or is the cornea truly unique in its sensitivity to belantamab mafodotin?

Asim V. Farooq, MD: One thing about the cornea is that we can see it. It’s right in front of us and we can examine it with a microscope. I don’t know that I would draw the conclusion that the corneal epithelium is necessarily very different in nature from other epithelial layers in the body or other cell types, except that it can be easily visualized and patients can potentially experience changes in their vision.

Sagar Lonial, MD: What we know from the adverse event profile of DREAMM-2 is that, beyond ocular toxicity, the most common adverse events were hematologic, which is not a surprise given that we are basically using a vinca-like compound in bone marrow. Beyond that, belantamab mafodotin actually is relatively easy to administer and I think that makes this an attractive therapy. Many of the treatments we use in refractory multiple myeloma require very frequent visits to the office and to the infusion center and the clinic for supportive care. This ADC does not require as much—for most patients, typically just 1 visit every 3 weeks, especially if they start with good blood counts at baseline. So I think, in that sense, other than the newness of the keratopathy piece, we really have not observed anything more than we would expect.

Can a patient receive all 3 types of BCMA-targeted therapy, or does one preclude another? If so, what therapy cannot follow the other?

Anthony Perissinotti, PharmD, BCOP: Currently, the only FDA-approved BCMA agent that we have is belantamab mafodotin. Any other BCMA-targeted therapy would only be available as part of a clinical trial. Currently, the answer to that would be dictated by the trial protocol eligibility criteria in terms of previous therapy. However, in terms of the pharmacology, all of these agents work very differently so one should not preclude another.

Sagar Lonial, MD: Initially, the trials excluded previous therapy with a BCMA-directed treatment. However, now many trials include separate cohorts to allow prior exposure to BCMA. Both the bispecific T-cell engager and the CAR T-cell trials are allowing this, presuming the previous BCMA-targeted therapy is belantamab mafodotin because it is approved by the FDA. There is some anecdotal experience of patients responding to a BCMA-targeted bispecific T-cell engager or bispecific antibody after belantamab mafodotin and BCMA expression is not lost as a consequence of progression. So, as of now, one BCMA therapy does not necessarily preclude another, but we await further data as the trials progress.