Visualization of Keratopathy Associated With the Antibody-Drug Conjugate Belantamab Mafodotin Using Infrared Imaging in Patients With Multiple Myeloma.

Belantamab mafodotin (Blenrep, GlaxoSmithKline Limited, Dublin, Ireland) is a first-in-class antibody–drug conjugate (ADC) introduced as a promising treatment option for patients with relapsed/refractory multiple myeloma (RRMM). 1 , 2 It is composed of a humanized Fc-modified anti-B-cell maturation antigen (BCMA) monoclonal antibody coupled with the cytotoxic agent monomethyl-auristatin-F (MMAF) through a noncleavable maleimidocaproyl linker. 3 Currently, Blenrep is being reevaluated for combination therapy with other agents as well as for earlier administration in several ongoing studies. 4

The development of superficial keratopathy is the most frequent adverse event associated with Blenrep treatment, reported in 72% of patients in the DREAMM-2 study. 5 Evidence suggests that the cytotoxic payload of an ADC is strongly associated with the corneal events observed. 6 Although keratopathy is generally reversible, 5 , 7 it still has a major impact on treatment outcomes, leading to significant dose reductions and delays in 23% and 47% of patients, respectively. 8 A multiinstitutional study by Mohan et al 9 on real-life experience regarding corneal toxicity in Blenrep treatment reported that 25% of patients discontinued treatment because of ocular side effects despite partial remission of RRMM.

Keratopathy-associated symptoms include blurred and reduced vision and discomfort due to dry eye disease. These symptoms can result in impaired quality of life and have led some patients to discontinue therapy despite a good response. 9 Moreover, progression-free survival was significantly worse in patients who discontinued treatment than in those who were able to restart Blenrep treatment. 9 Therefore, the effective identification and management of Blenrep-associated keratopathy are crucial to support patients’ quality of life, prevent treatment interruptions, and ensure treatment success.

Blenrep-associated keratopathy is diagnosed by slit-lamp examination and characterized by the presence of microcyst-like epithelial changes (MECs). To assess ocular side effects, the Keratopathy and Visual Acuity (KVA) scale was established within the DREAMM-2 protocol. 5

The KVA scale currently serves as the sole foundation for issuing recommendations of dose reduction, delay, or discontinuation in response to ocular side effects. However, this approach largely relies on the subjective assessment of keratopathy by an ophthalmologist, as the only criteria considered by the KVA scale are the extent of keratopathy subjectively graded on slit-lamp examination and the change in best corrected visual acuity (BCVA) from baseline.

Therefore, multimodal corneal imaging techniques to enhance the visualization of Blenrep-associated keratopathy and identify possible objective imaging biomarkers are paramount. In few individual cases, specific corneal imaging techniques, including in vivo confocal microscopy (IVCM), anterior segment optical coherence tomography (AS-OCT), and corneal topography, have been used to characterize corneal changes. 5 , 10 – 15 In this study, we analyzed the use of anterior segment infrared reflectance (AS-IR) imaging of the cornea as a diagnostic tool to visualize Blenrep-associated keratopathy, with the potential to be incorporated into the current clinical workup and provide additional insights.

MATERIALS AND METHODS

Patient Recruitment and Treatment Protocol

This retrospective study was conducted at the University Hospital Schleswig-Holstein Campus Kiel and approved by the Institutional Review Board. All patients with RRMM who received Blenrep treatment between September 2020 and October 2021 were included in this study. Data were extracted by reviewing electronic and analog medical records as well as the ophthalmological imaging system database. Patients received Blenrep at an initial dose of 2.5 mg/kg, subsequently administered every 21 days unless deemed ineligible because of adverse events (AEs). In these cases, the dosage was reduced to 1.9 mg/kg or administration was delayed, according to the GSK guidelines. 16

Ophthalmological Examinations

All patients underwent baseline ophthalmological evaluation before the first dose and serially before every planned dose administration. Ophthalmological evaluation included mandatory determination of best corrected visual acuity (BCVA) and slit-lamp examination of the anterior segment. Blenrep-associated superficial keratopathy was graded using the KVA scale. 5

Imaging Techniques

Specific corneal imaging using AS-IR of the cornea (Heidelberg Engineering Spectralis HRA + OCT, Heidelberg Engineering, Germany, Anterior Segment Module λ = 815 nm, 30 × 30-degree field of view), AS-OCT, and slit-lamp photography were performed.

Evaluation Technique of AS-IR Images

AS-IR images were used to evaluate visible corneal alterations, as illustrated in Figure 1 . For analysis, the corneal surface was sectioned into three zones (central 4 mm, paracentral 8 mm, and peripheral >8 mm in diameter from the center point of the cornea) and subsequently divided into four quadrants (Fig. 1 , left side). The appearance of corneal alterations was graded on a scale of 0 to 4 points within each quadrant, and a cumulative score was calculated per zone for each eye (Fig. 1 , right side). For statistical analysis, BCVA was converted to LogMAR and Pearson correlation coefficients were determined using rStudio software [R Core Team (2021). R: Language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL: https://www.R-project.org/ ].

RESULTS

Nine patients were included in this study (Table 1 ). The mean age of the patients was 62 ± 7 years; five patients were female, and four were male. The BCVA at baseline ranged from 0.7 to 1.25 decimal. Two patients had mild superficial keratopathy before Blenrep treatment initiation: one due to dry eye disease and one secondary to lagophthalmos due to Bell's palsy. The follow-up period was between three and 57 weeks, with a mean of 6 ± 5 Blenrep doses administered during this period. Seven patients underwent a dose reduction or delay. In one patient, treatment was discontinued because of severe symptomatic ocular adverse effects.

Seven patients developed Blenrep-associated kerato-pathy, initially presenting between 3 and 6 weeks after the first-dose application. Four patients developed grade 2 kerato-pathy, which became evident after the third-dose application in three cases. One patient developed severe keratopathy, which ultimately progressed to grade 4 keratopathy.

On AS-IR, Blenrep-associated corneal alterations present as hyporeflective lesions of varying size and density. These lesions were visible in all but one patient at the follow-up after the first Blenrep infusion. In most patients, AS-IR showed small, loosely grouped pinpoint lesions distributed in a semicircular pattern predominantly located in the paracentral cornea and resembling the distribution of MECs on slit-lamp examination (Fig. 2 ).

In more severe cases, the lesions appeared as confluent patches on AS-IR. In one patient, an influx of densely packed hyporeflective signal alterations exhibiting a whorl-like pattern involving the paracentral and peripheral cornea bordering the limbus was identified on AS-IR following the second Blenrep dose (Figs. 3 A, B). At this point, visual acuity remained unaffected. However, despite the interim attenuation of MECs following dose reduction, persistent epithelial defects developed in the long term (Figs. 3 E, 4 B). Ultimately, the sustained ocular side effects led the patient to discontinue the Blenrep treatment. By contrast, the patients in our study who showed only slight alterations in BCVA and were able to maintain regular dosing schedules, albeit at a reduced dose, displayed a more dispersed, circular distribution of AS-IR lesions without patchlike involvement of the peripheral or limbal regions at any time.

AS-IR images showed dynamic changes in the distribution, amount, size, and density of corneal alterations between follow-ups. The lesions tended to become more dispersed and smaller toward the central cornea. This was particularly apparent in the preceding dose reductions and treatment interruptions (Fig. 5 ). In one patient, AS-IR was obtained at weekly intervals between two Blenrep administrations (Fig. 5 A–C) during a 3-week treatment delay before treatment was resumed at a reduced dose (Fig. 5 D). Here, consistent regression of the hyporeflective lesions regarding size and density, as well as their migration toward the central cornea, could be observed (Fig. 5 E–F). Following treatment termination, lesions visible on infrared imaging receded with no new influx of AS-IR lesions in more peripheral areas (Figs. 5 G–H). Correspondingly, slit-lamp examination showed complete remission of Blenrep-associated keratopathy by week 57.

We examined the possible correlation between the loss of BCVA compared with baseline and scoring based on the AS-IR images. While no significant correlation was found between the central and peripheral corneal zones (Figs. 6 A, C), regression analysis suggested a positive correlation between the cumulative keratopathy score on AS-IR images of the paracentral corneal zone and BCVA loss compared with baseline (r = 0.29779, P < 0.001, Fig. 6 B).

DISCUSSION

Identifying and managing Blenrep-associated kerato-pathy effectively to prevent more severe courses and dose interruptions are crucial for overall treatment success. As the only effective available tools to mitigate ocular complications are dose modifications, such as dose reductions or treatment delays, the assessment of individual risk and tolerance to ocular adverse events is vital. Several trials have investigated different treatment combinations, dosing schedules, and frequency schedules to optimize the risk-benefit profile, 17 – 19 efforts which are ongoing. Objective documentation provided by specific corneal imaging to monitor ocular findings in individual patients would aid in earlier detection of ocular lesions and subsequent fine-tuning of the treatment protocol. However, current ophthalmological examination protocols do not include multimodal imaging methods or photograph documentation, which hampers the data analysis.

Owing to the small size of MECs, capturing them using slit-lamp photography requires special consideration. 5 Although IVCM is very sensitive for identifying potential preclinical findings, its small field of view limits its use in monitoring corneal changes. By contrast, AS-IR provides a quick, noncontact imaging technique for generating an overview of the corneal surface, whereby its entirety can be assessed using only one image. This enables the examiner to accurately detect small fluctuations between visits and chronicle keratopathy progression and to effectively monitor its activity over extended treatment durations. Based on our experience, the ability to evaluate the degree of keratopathy at a glance and perform side-by-side comparisons between visits is highly practical in an everyday clinical setting and can be readily integrated into existing diagnostic workups.

AS-IR may be superior to slit-lamp examination in demonstrating early manifestations of Blenrep-associated keratopathy. AS-IR corneal changes in Blenrep-associated keratopathy present as distinct hyporeflective dots of variable size, predominantly located in the paracentral zone. Corresponding AS-OCT showed a localized increase in heterogeneous signal intensity with hyperreflective bands visible within the corneal epithelium and slight signal disruption of the underlying Bowman layer (Fig. 4 ), as characterized in previous reports. 12 – 14 While the spatial distribution of AS-IR lesions resembled that of MECs observed on slit-lamp examination, hyporeflective dots were visible on AS-IR up to 6 weeks before MECs were detected by the examiner on slit-lamp examination.

On AS-IR, the increased contrast between dark corneal spots against the brighter backdrop of the iris can be exploited to demarcate corneal changes, making it more sensitive to discrete alterations that can be difficult to appreciate on slit-lamp examination and, thus, risk going unnoticed. Correspondingly, AS-IR lesions are more difficult to discern against the dark backdrop of the pupil, which can limit the capacity to evaluate changes in the optical zone within the central cornea depending on pupil size.

AS-IR enables the semiquantitative analysis of kerato-pathy activity, as illustrated by our proposed scoring system. Previous studies have not established a linear relationship between keratopathy severity and vision loss. In our study, in contrast to the cumulative keratopathy scores calculated for the central and peripheral corneal zones, the scores calculated for the paracentral cornea seemed to correlate with BCVA loss from the baseline values. Higher scores for the paracentral cornea coincided with a decline in visual acuity compared with baseline. However, keratopathy scores calculated for the central corneal zone may be affected by the abovementioned limited visibility of AS-IR lesions in the optical zone, which warrants prudence when considering the lack of correlation regarding BCVA loss from baseline.

Interestingly, refractive shifts and changes in corneal curvature based on the localization of MECs induced by antibody–drug conjugates have been reported by Canestro et al, whereby peripheral MECs were associated with hyperopic shifts and paracentral/central MECs were associated with myopic shifts and corneal steepening. 20

In a case reported by Marquant et al, density of “bunch-of-grape” lesions in the basal epithelium observed on IVCM related to vision loss, rather than the presence of visible MECs 11 themselves. IVCM seemed to illustrate Blenrep-containing cells at various stages of induced apoptosis, suggesting that MECs may be interpreted as degenerative epithelial cysts that evolved from apoptotic epithelial cells. 10 It has been hypothesized that the cornea is exposed to Blenrep through limbal vessels or tears and subsequently internalized through micropinocytosis into basal epithelial cells. During migration of epithelial cells from the basal layer in the periphery toward the surface and center of the cornea, apoptosis is induced. 5 AS-IR changes emulated the space–time kinetics of keratopathy activity. Following dose reduction, dose delay, or treatment discontinuation, regression of hyporeflective lesions was observed over time. Lesions migrated toward the central cornea with reduced influx from more peripheral areas, suggestive of a “washout” phenomenon consistent with the superficial shedding of apoptotic cells by desquamation. 7

Our analysis of the belantamab mafodotin compound by infrared spectrometry did not show absorption at diagnostic AS-IR wavelengths (not shown). AS-IR lesions may, therefore, represent the aforementioned preapoptotic or degenerated epithelial cells themselves or an accumulation of apoptotic cell products in the intercellular space, as previously visualized using IVCM and AS-OCT. 10 – 14 AS-IR lesions may constitute precursors of MECs and reveal keratopathy at an early stage.

In addition, the consideration of the location, arrangement, and extension of AS-IR lesions into certain corneal zones may aid in anticipating severe corneal side effects. This can facilitate earlier treatment adjustments aimed at enhancing tolerability and preventing long-term treatment discontinuation. In our study, patients who displayed high tolerability of keratopathy and were able to maintain a regular dosing schedule showed only mild fluctuations in the number, density, and dispersion of AS-IR lesions, which seemed to develop a steady state. The lesions tended to adopt a halo-shaped pattern with few areas of confluence and a notable lack of peripheral and limbal involvement.

By contrast, the only patient in our study who developed severe keratopathy resulting in treatment discontinuation presented with a whorl-like appearance of AS-IR lesions involving the peripheral and limbal cornea following the second Blenrep infusion yet before significant visual loss. Severe keratopathy and visual loss did not manifest until 9 weeks and corneal ulceration until 12 weeks later. Bausell et al described whorl-like staining patterns on slit-lamp examination reminiscent of limbal stem cell dysfunction in 67% of the patients. However, these changes took an average of 80 days to become apparent. 21 AS-IR seems to allow early visualization of such changes. Whorl-like patterns have been attributed to the centripetal movement of corneal epithelial cell migration and architecture of the subbasal nerve plexus. 22 The specific whorl-like constellation exhibited on AS-IR may also be interpreted as an indicator of the recently suggested neurotoxic effect of Blenrep, denoted by damage and fragmentation to the subbasal nerve plexus on IVCM, as reported by Aschauer et al. 10

To the best of our knowledge, this is the first longitudinal study to evaluate the application of AS-IR imaging in visualizing Blenrep-associated keratopathy. Previous reports are limited to one study mentioning AS-IR corneal imaging regarding corneal toxicity of mirvetuximab soravtansine, an ADC used to treat ovarian cancer, where five patients were monitored using AS-OCT and hyporeflective dots were detected on AS-IR images. 23

The OCT anterior segment module used in this study includes the infrared filter in its hardware and is widely commercially available, thus making AS-IR image acquisition accessible for use in real-life settings and our approach reproducible in different facilities. This would permit the exchange and comparison of images between different centers, which has significant implications for clinical trials and practice. Recently, multiple reviews have focused on the adverse effects of ADCs on the ocular surface, as observed with a number of newly developed ADCs against different cancers. 24 , 25 It was concluded that there is a need for less variability of interpersonal conditions within studies and implementation of diagnostic methods to enable more powerful analyses. 25 The addition of AS-IR imaging as a diagnostic monitoring tool may be pertinent for addressing this issue. Additional studies are warranted to optimize AS-IR scoring systems and identify precise cutoffs to determine the thresholds of tolerable ocular side effects.

In conclusion, infrared imaging can objectively demonstrate keratopathy activity and allow for the analysis of space–time kinetics. The findings were consistent with keratopathy detected on slit-lamp examination and related to the Blenrep infusion schedule, dose alterations, and drug-free interval periods. In multiple patients, AS-IR lesions were identified before MECs were noted on slit-lamp examination. Using AS-IR imaging may, therefore, enable earlier detection of keratopathy and, in turn, could contribute to a more proactive approach to mitigation strategies, potentially preventing more severe ocular adverse effects with subsequent treatment discontinuation.

Overall, AS-IR imaging may be a useful complementary imaging tool to evaluate Blenrep-associated keratopathy. The incorporation of routine AS-IR imaging into the diagnostic workup of patients treated with Blenrep seems reasonable and realistic in an everyday clinical setting.

Footnotes

The authors have no funding or conflicts of interest to disclose.

Contributor Information

Bernhard Nölle, Email: Bernhard.Noelle@uksh.de.

Natalie Schub, Email: Natalie.Schub@uksh.de.

Johann Roider, Email: Johann.Roider@uksh.de.

REFERENCES

1. Trudel S, Lendvai N, Popat R, et al. Targeting B-cell maturation antigen with GSK2857916 antibody-drug conjugate in relapsed or refractory multiple myeloma (BMA117159): a dose escalation and expansion phase 1 trial. Lancet Oncol. 2018;19:1641–1653.

2. Lonial S, Lee HC, Badros A, et al. Belantamab mafodotin for relapsed or refractory multiple myeloma (DREAMM-2): a two-arm, randomised, open-label, phase 2 study. Lancet Oncol. 2020;21:207–221.

3. Tai YT, Mayes PA, Acharya C, et al. Novel anti–B-cell maturation antigen antibody-drug conjugate (GSK2857916) selectively induces killing of multiple myeloma. Blood. 2014;123:3128–3138.

4. Morè S, Corvatta L, Manieri VM, et al. Novel immunotherapies and combinations: the future landscape of multiple myeloma treatment. Pharmaceuticals. 2023;16:1628.

5. Farooq AV, Degli Esposti S, Popat R, et al. Corneal epithelial findings in patients with multiple myeloma treated with antibody-drug conjugate belantamab mafodotin in the pivotal, randomized, DREAMM-2 study. Ophthalmol Ther. 2020;9:889–911.

6. Donaghy H. Effects of antibody, drug and linker on the preclinical and clinical toxicities of antibody-drug conjugates. mAbs. 2016;8:659–671.

7. Eaton JS, Miller PE, Mannis MJ, et al. Ocular adverse events associated with AntibodyDrug conjugates in human clinical trials. J Ocul Pharmacol Ther. 2015;31:589–604.

8. Lonial S, Nooka AK, Thulasi P, et al. Management of belantamab mafodotin-associated corneal events in patients with relapsed or refractory multiple myeloma (RRMM). Blood Cancer J. 2021;11:103.

9. Mohan M, Rein LE, Thalambedu N, et al. Corneal toxicity with belantamab mafodotin: multi-institutional real-life experience. Am J Hematol. 2022;97:E451–E453.

10. Aschauer J, Donner R, Lammer J, et al. Corneal toxicity associated with belantamab mafodotin is not restricted to the epithelium: neuropathy studied with confocal microscopy. Am J Ophthalmol. 2022;242:116–124.

11. Marquant K, Quinquenel A, Arndt C, et al. Corneal in vivo confocal microscopy to detect belantamab mafodotin-induced ocular toxicity early and adjust the dose accordingly: a case report. J Hematol Oncol. 2021;14:159.

12. Mencucci R, Cennamo M, Alonzo L, et al. Corneal findings associated to BelantamabMafodotin (Belamaf) use in a series of patients examined longitudinally by means of advanced corneal imaging. J Clin Med. 2022;11:2884.

13. Matsumiya W, Karaca I, Ghoraba H, et al. Structural changes of corneal epithelium in belantamab-associated superficial keratopathy using anterior segment optical coherence tomography. Am J Ophthalmol Case Rep. 2021;23:101133.

14. Chuang K, Pineda R, Liu S. Belantamab mafodotin associated corneal microcyst-like epithelial changes. Am J Ophthalmol Case Rep. 2022;25:101392.

15. Mickevicius T, Pink AE, Bhogal M, et al. Dupilumab-induced, tralokinumab-induced, and belantamab mafodotin–induced adverse ocular events—incidence, etiology, and management. Cornea. 2023;42:507–519.

16. Blenrep-epar-product-information_en.pdf. https://www.ema.europa.eu/en/documents/product-information/blenrep-eparproduct-information_en.pdf . Accessed July 12, 2023.

17. Iida S, Sunami K, Mishima Y, et al. Safety, pharmacokinetics, and efficacy of belantamab mafodotin monotherapy in Japanese patients with relapsed or refractory multiple myeloma: DREAMM-11. Int J Hematol. 2023;118:596–608.

18. Hultcrantz M, Kleinman D, Ghataorhe P, et al. Exploring alternative dosing regimens of single-agent belantamab mafodotin on safety and efficacy in patients with relapsed or refractory multiple myeloma: DREAMM-14. J Clin Oncol. 2022;40:TPS8073.

19. Avivi I, Shragai T, Luttwak E, et al. Treatment with low-dose, single-agent belantamab mafodotin is safe and provides long-term responses in heavily pretreated multiple myeloma patients. Eur J Haematol. 2024;112:367–370.

20. Canestraro J, Hultcrantz M, Modi S, et al. Refractive shifts and changes in corneal curvature associated with antibody-drug conjugates. Cornea. 2022;41:792–801.

21. Bausell RB, Soleimani A, Vinnett A, et al. Corneal changes after belantamab mafodotin in multiple myeloma patients. Eye Contact Lens. 2021;47:362–365.

22. Patel DV, McGhee CNJ. Techniques for wide-field assessment of the human corneal subbasal nerve plexus. Cornea. 2013;32:e140–e141.

23. Corbelli E, Miserocchi E, Marchese A, et al. Ocular toxicity of mirvetuximab. Cornea. 2019;38:229–232.

24. Boucher R, Haigh O, Barreau E, et al. Ocular surface toxicities associated with modern anticancer therapies. Surv Ophthalmol. 2024;69:198–210.

25. Domínguez-Llamas S, Caro-Magdaleno M, Mataix-Albert B, et al. Adverse events of antibody–drug conjugates on the ocular surface in cancer therapy. Clin Transl Oncol. 2023;25:3086–3100.