Introduction

Eyelash megalotrichosis (MT) is an overgrowth of eyelashes [1]. This rare phenomenon can be a manifestation of congenital syndromes and diseases, or acquired condition secondary to malignancy, infection (i.e., HIV), and autoimmune diseases, or side effect of therapy with several pharmaceutical agents [2,3]. MT was reported as a side effect of treatment with Prostaglandin analogues (PGs) and Tyrosine Kinase Inhibitors (TKIs) [1]. PGs are a group of agents used as topical ophthalmologic treatments (drops) for glaucoma that reduce intra-ocular pressure (IOP) [4]. PGs-induced MT is a common and well-documented side effect. The PGF2α-analogue bimatoprost increases the number of eyelashes and induces hair shafts thickening and elongation in murine models [5]. In human, 26%-76% of glaucoma patients treated with the PGs latanoprost, bimatoprost or travoprost experience eyelash growth, per different agents [4]. TKIs are small molecules which inhibit growth factor receptors pathways, epidermal mainly, and are widely used in several cancers therapy with non-small cell lung cancer (NSCLC) being the most prominent [6]. These agents have several dermatologic and ophthalmologic side effects, which among others include MT [6]. TKIs-induced MT was reported among patients treated with erlotinib [7,8] and gefitinib [9]. it was also reported following administration of monoclonal antibodies targeting epidermal growth factor receptors (EGFR) as cetuximab and panitumumab [10,11]. A retrospective study on ocular toxicities among 69 patients treated with EGFR inhibition for malignancy found that 22 patients developed MT or trichiasis in 12 months, of which 9 (13%) required removal or trimming of eyelashes due to irritation and 2 (2.9%) suffered corneal abrasion [12]. Of 42 patients treated with the TKI erlotinib, 12 (28.6%) developed MT or trichiasis (growth of lashes towards the eye) [12]. Thus, MT might be a disturbing side effect not only cosmetically, but also due to risk for corneal irritation, abrasion, and the need for consequent ophthalmologic intervention [12]. TKIs-induced MT was mainly described in discrete case reports [7–11]; As both PGs-treated glaucoma and TKIs-treated cancers are common health conditions [4,12], substantial portion of patients will develop MT as an adverse effect. Our study aims to improve clinical and morphological recognition of different forms of MT, and thus improve awareness and clinical care of patients with this condition.

Methods

Results

Ophthalmologic evaluation

All Patients were evaluated by a specialist in ocular oncology from the Ocular Oncology service, The Goldschleger eye institute, Sheba medical center. Visual acuity (VA) and Intra-Ocular Pressure (IOP) did not show any significant differences between the two groups; pupil and iris examination were normal for all patients. Since the PGs glaucoma patients were referred due to suspected or diagnosed malignant ocular lesions, major differences exist between the two groups regarding fundus examination (not shown). Higher proportion of TKIs-treated patients presented with MT-associated complaints (3/5 Vs. 0/6 patients in the PGs-group, p = .06). However, no significant difference was found in proportion of patients needed lubrication therapy (2/5 TKIs-treated Vs. 1/6 PG-treated, p = .55).

Table 1 summarizes patient’s characteristics among PGs and TKIs groups and demonstrates that no significant differences were found between the groups regarding demographic, clinical or ophthalmologic exams, apart from MT-related complaints and need for therapy.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Patients’ characteristics.

https://doi.org/10.1371/journal.pone.0313326.t001

Morphological assessment of MT

For each patient, assessment of MT morphological features was conducted independently by VD and RG. The features assessed on morphological examination of MT included poliosis, trichiasis, number of eyelash rows on LL and UL, individual elongated eyelash (defined as ≥2 times the length of regular eyelash) and eyelash curvature based on the detailed description in the medical record and/or slit lamp photographs. Representing pictures, with prominent morphological features are depicted in Fig 1. TKIs-treated cancer patients had significantly more individual elongated eyelashes (4/4 on TKIs group Vs. 1/6 on PGs group, Fisher’s exact p = .047) and number of UL eyelash rows (medians: 5.5 on TKIs group Vs. 4 on PGs group, Wilcoxon rank-sum p = .03). Trends toward significance were found regarding higher poliosis proportion among PGs-treated patients (0/4 on TKIs group Vs. 4/6 on PGs group, Fisher’s exact p = .076) and lower proportion of eyelash curvature among PGs-treated patients (4/4 on TKIs group Vs. 2/6 on PGs group, Fisher’s exact p = .076). Trichiasis and number of LL eyelash rows did not differ between the two groups. Table 2 summarizes morphological features comparison between the two groups and time from drug introduction to MT presentation.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1.

Morphological features of (A) TKIs-induced and (B) PGs-induced MT. (A) representative features of TKIs-induced MT (blue arrows): (1) elongated individual eyelashes with curvature (2) increased UL rows of elongated lashes (3) elongated, curved lashes with trichiasis (4) increased UL eyelashes rows; (B) representative features of PGs-induced MT (green arrows) (1) non-curved elongated lashes (2) non-curved lashes with poliosis (depigmented lashes, arrows) and trichiasis (arrowhead) (3) poliosis (arrows) and trichiasis (arrowheads). TKIs–Tyrosine Kinase Inhibitors; PGs–Prostaglandin analogues; MT–Megalotrichosis; UL–Upper Lid.

https://doi.org/10.1371/journal.pone.0313326.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Comparison of MT morphological features and drug-MT interval between patients with PGs-treated glaucoma and patients with TKI-treated malignancy.

https://doi.org/10.1371/journal.pone.0313326.t002

Discussion

Glaucoma treated with PGs and cancer treated with TKIs are common health conditions; Up to 76% of glaucoma patients treated with PGs [4] and 28.6% of patients with TKIs-treated malignancy [12] might develop MT as a side effect of their therapy. While MT is a common, well-documented, sometimes desired side effect of PGs treatment with minor clinical significance [13], clinical effects of TKIs-induced MT were not systematically evaluated and as this study suggests it might be of more clinical importance. In our center, 11 patients were referred for evaluation in the Ocular Oncology service of suspected conditions over a period of 10 years and were found to suffer of MT. The discrepancy between the high reported rates of MT among PGs or TKIs-treated patients and the low numbers of patients in our series might represent referral bias and possibly imply that most MT cases are asymptomatic and are diagnosed as accidental finding or have mild symptoms treated in primary care centers. Another potential limitation of the study is that glaucoma patients were referred to ocular oncology department due to background non-ocular malignancy suspected of dissemination; thus, they may not well represent the general population of patients with glaucoma.

Although reports on the prevalence and incidence of MT developing as side effect of PGs or TKIs treatment are scarce in the literature, this study focused on characterizing disparities between these two forms of MT, and due to above mentioned limitations, describing the epidemiological features of this phenomenon was not part of this study.

A previous study analyzed ocular toxicities among patients treated with EGFR inhibition who were referred to the cornea service, over a period of 5 years [12]. In total, 22 patients presented with MT or trichiasis; nonetheless, only 9/22 (40.9%) patients required intervention, a proportion similar to our 2/5 (40%) patients who received therapy for TKIs-induced MT. Similar data regarding PGs-induced MT complaints or need for treatment were not found by the authors, hence comparison was not available. However, in our cohort 3/5 patients treated with TKIs had MT-related complaints, as opposed to 0/6 patients treated with PGs. This finding may imply that TKIs-induced MT might be more disturbing to the patient than PGs-induced MT. Interventions reported in the literature for TKIs-induced MT include topical lubrication agents and eyelash trimming in extreme cases [12]. In our experience, 2 patients received topical preservative-free vitamin A-containing lubrication ointment, and 1 of these patients also needed eyelash trimming to alleviate corneal irritation due to MT.

Despite our small sample sizes, significant or near-significant differences distinguish each form of MT: TKIs-induced MT is characterized by more individual elongated eyelashes, higher number of UL eyelashes rows, and eyelash curvature; PGs-induced MT had higher proportion of poliosis. While to our knowledge reports on morphological features of TKIs-induced MT are absent in the literature, Johnstone described hypertrichosis among 43 patients treated with PGs (latanoprost) drops [14]; Concordant with our work, he noted increased eyelash rows and eyelash curvature as morphological features of latanoprost-induced MT. The study also noted increased eyelash pigmentation as a side effect; opposingly, and compatible with our findings, Chen et al. reported 7 cases of poliosis following topical Prostaglandin F2-α (PGF2α) administration for primary open-angle glaucoma [15].

In our study, time from drug administration to MT onset was 440 days [range:155–878] among PGs-treated group and 176 days [56–418] among TKIs-treated group. The difference was not found statistically significant, probably due to small cohort. Larger cohorts are needed to validate this trend. Studies documenting the time interval from drug to MT onset are scarce. Patients at Johnstone’s cohort were treated with latanoprost for at least 10 weeks (mean:19.8, range:11–40) before evaluating latanoprost-induced MT [14], yet time to onset is not specifically documented. A paper reviewing several clinical trials on PGs for treating glaucoma reported eyelash growth as adverse effect 3, 6, 9, and 12 months after starting PGs therapy [4], yet no case was reported among 1-month trials. Although not stated by authors, MT was more prevalent with longer duration of use [4]. Consistent with our results, A study describing ocular toxicities of EGFR inhibition in patients with malignancy noted that all cases of trichomegaly occurred within 3–5 months after starting therapy [12]. However, trichomegaly and trichiasis were reported simultaneously, and times to MT were not stratified to therapy as both TKIs and monoclonal antibodies were used to inhibit EGFR. Consequently, time interval from TKIs administration to MT onset was not reported.

Several other medical conditions (e.g. congenital syndromes, autoimmune diseases, HIV infection) and therapeutic agents (e.g. Calcineurin inhibitors, topiramate) are reported in the literature as potential inducers of MT [1,13]. However, PGs and TKIs induced MT have the highest quality support in literature while MT association with other medical conditions or therapies have lower quality of evidence [13]. In our cohort, over 10 years of follow-up, 11 cases of MT were diagnosed and treated following the use of PGs or TKIs. None of the patients had other medical conditions or therapies suspected to induce MT. Hence, despite potential clinical relevance, the features of MT induced by other medical conditions and medications are beyond the scope of this study.