Ocular Surface Squamous Neoplasia: Changes in the Standard of Care 2003 to 2022

Abstract

Purpose:

To elucidate treatment preferences for ocular surface squamous neoplasia (OSSN) and to examine the changes in treatment modalities over the past two decades.

Methods:

An electronic survey was distributed to members of the Cornea Society, Ocular Microbiology and Immunology Group, and 4 international cornea specialist listservs. Questions examined medical and surgical treatment preferences, and results were compared to surveys administered in 2003 and 2012.

Results:

A total of 285 individuals responded to the survey; 90% of respondents were self-classified as cornea specialists. Seventy three percent reported using primary topical monotherapy to treat OSSN as compared to 58% in 2012 (p=0.008). Compared to 2003, the percentage use of topical interferon significantly increased (p<0.0001) from 14% to 55%, 5-fluorouracil increased (p<0.0001) from 5% to 23%, and mitomycin C decreased (p<0.0001) from 76% to 19% as a primary monotherapy. The frequency of performing excision without the use of postoperative adjunctive medical therapy decreased significantly (p<0.0001), from 66% to 26% for lesions less than 2 millimeters, 64% to 12% for lesions between 2 and 8 millimeters, and 47% to 5% for lesions greater than 8 millimeters from 2003 to 2022. More clinicians initiated topical immuno/chemotherapy without performing a biopsy as compared to 2003 (31% vs 11%, p<0.0001).

Conclusion:

These results demonstrate a paradigm shift in the management of OSSN. The use of primary medical therapy as a first approach has significantly increased, with a reduction in the frequency of performing surgical excision alone.

Introduction

Ocular surface squamous neoplasia (OSSN) is an umbrella term used to identify a mostly non-pigmented cancerous epithelial abnormality of the conjunctiva and cornea. OSSN comprises a wide spectrum of histologic diagnoses, ranging from mild epithelial dysplasia to invasive squamous carcinoma.¹ Incidence of OSSN is approximately 0.03-1.9 per 100,000 persons in the United States (US) and Australia, with the lower estimate reflecting a histologic diagnosis of squamous cell carcinoma alone and the higher estimate including dysplasia, carcinoma in situ, and invasive squamous cell carcinoma.^{2, 3} In African populations, the incidence of OSSN has been estimated between 1.6-3.4 per 100,000,³ with the higher estimates compared to US and Australia thought to be driven by human immunodeficiency virus (HIV) associated immunosuppression.^{4, 5} The leading risk factors for OSSN are exposure to ultraviolet (UV) radiation, immunosuppression, and human papilloma virus (HPV) infection, but the frequency of these risk factors likely vary by study population.⁶

Traditionally, surgical excision dominated as the method of treatment for OSSN. However, in response to its potential downsides (e.g., risk of recurrence, scarring, limbal stem cell deficiency), several alternative therapies have emerged. The principal agents include, 5-fluorouracil (5FU), interferon-α2b (IFN), and mitomycin C (MMC). 5FU, an agent that disrupts DNA replication and cell growth by blocking thymidylate synthase⁷, was first used topically to treat OSSN in 1986 and its use has resurfaced and increased during the 21^st century.⁸ Benefits of 5FU include high resolution rates (82-100%) and low recurrence (10-14%) frequencies.^9-12 IFN, a protein with anti-proliferative, immunomodulatory, and antiviral effects¹³, was first used against OSSN in 1994¹⁴ and has since demonstrated efficacy both as a topical eye drop and as perilesional injection.^{15, 16} Benefits of IFN include high resolution rates (75-100%) and low recurrence (0-25%) frequencies.^17-19 MMC, an antimetabolite isolated from Streptomyces caespitosus bacteria²⁰, binds to DNA to inhibit DNA synthesis.^{21, 22} Topical MMC was first used against OSSN in 1994²³ with high resolution (76-100%) and low recurrence (0-20%) frequencies reported.^24-31 A downside of MMC, compared to 5FU and IFN, is that its side effect profile is less favorable, with pain and redness after use, and the potential for long term limbal stem cell deficiency, melts, and punctal stenosis. Combination therapy, consisting of both surgical excision followed by adjuvant medical intervention, has also been used, with decreased recurrence rates compared to surgical excision alone.³²

In 2003, Stone et al. sent a survey to all members of the Cornea Society and Ocular Microbiology and Immunology Group (OMIG) to report practice patterns for OSSN management.³³ The questionnaire focused on the preferred method of therapy to treat OSSN of various sizes. Nine years later (in 2012), Adler et al. sent the same questionnaire to members of the Cornea Society and OMIG.³⁴ New data from the survey was compared to Stone et al.’s results from 2003. Overall, there was a noticeable increase in topical therapy usage, specifically IFN (preferred by 58% of clinicians in 2012 compared to 14% in 2003) and a concomitant decline in surgery as monotherapy.³⁴ The rates of surgical excision alone, without adjunctive post-operative topical medical therapy, decreased from 66% to 51%, 64% to 25%, and 47% to 15% for lesion sizes less than 2mm, between 2 and 8 mm, and greater than 8 mm from 2003 to 2012, respectively.³⁴

Since 2012, shifts in the practice patterns have emerged in response to advancements in diagnostic technology such as high-resolution anterior segment optical coherence tomography, impression cytology, and refinement in surgical technique. The purpose of this paper was to evaluate standards in the treatment of OSSN given these developments, to extend the reach of the survey to include both national and international expert clinicians, and to compare these results to prior studies.

Materials and Methods

This study received exemption from the Institutional Review Board from the University of Miami. A survey was created through the online survey tool, SurveyMonkey (www.surveymonkey.com). The survey contained 24 questions that examined participants’ medical treatment preferences, surgical approaches, and current access to therapies for the management of OSSN. Of these, 11 questions were identical to those asked in 2003 and 2012, and 13 were novel. Novel questions addressed the number of years in practice since training, practice location, practice patterns, OSSN treatment regimens, drug availability, and factors associated with drug choice (see appendix). As with prior surveys, invitations to the electronic survey were sent to the electronic mailing lists of The Cornea Society, OMIG, and the online link to access the survey was disseminated during academic presentations at these conferences. In order to expand the international distribution, the survey was also sent to the listserv members of the KeraNet, Pancornea, and cornea specialist groups in India, Thailand, and Japan. Data from survey participants was collected and analyzed using the MedCalc Statistical Software version 20.115 (MedCalc Software, Ostend, Belgium; http://www.medcalc.org). The main outcome was treatment preferences reported in frequencies, using the comparisons of proportions test (Z test) to compare frequencies between the current and previous surveys (2003 and 2012). A p-value of less than 0.05 was considered statistically significant. This research study adhered to the tenets of the Declaration of Helsinki.

Results

Two-hundred and eighty-five participants responded to the survey. Ninety percent of the respondents considered themselves to be cornea specialists; the characteristics of the survey participants are summarized in Table 1. Approximately 51% of the survey participants were in practice after training for over 15 years and 47% of the participants practiced in the United States or Canada. The number of OSSN cases treated per year was variable, with 39% of clinicians indicated that they treated 0 - 5 cases per year and 33% between 5 -10 cases per year.

Table 1:

Characteristics of the survey participants

Years in Practice After Training	Frequency (%)
Greater than 15 Years	51
6-15 Years	32
5 Years or Less	16
Geographic Distribution
United States or Canada	47
South or Central America	26
Europe	11
Other	9
India	5
Asia	3
Number of OSSN Cases Seen Per-Year
Less than 5 Cases	39
5-10 Cases	33
10-15 Cases	15
15-20 Cases	8
20-30 Cases	1
Greater than 30 Cases	3
Specialty Training in Cornea*
Yes	90
No	10

OSSN- Ocular Surface Squamous Neoplasia, *Self-reported specialty training

Participants were surveyed regarding the frequency with which they use topical chemotherapy agents as monotherapy. Seventy three percent reported using primary topical monotherapy to treat OSSN as compared to 58% in 2012 (p=0.008). Preferred first-line agents included topical IFN (55%), 5FU (23%), and MMC (19%) eyedrops. The proportion of participants using 5FU as a first-line agent significantly increased (P<0.001) from 5% in 2003 to 23% in the current study. Among clinicians who used primary medical therapy, 79%, 77%, and 62% of respondents reported that they were able to access MMC, 5FU, and IFN, respectively at the time of the survey in 2021. If medical therapy was unsuccessful, the most utilized approach was surgical excision with topical adjunctive therapy (73%), whereas less favored methods included undergoing excision alone (15%) and administering a second topical therapy (12%). In the 2012 survey, 89% said they would perform surgical excision if primary medical therapy was unsuccessful. More clinicians initiated topical immuno/chemotherapy without performing a biopsy as compared to 2003 (31% vs 11%, p<0.0001). Table 2 compares responses to survey questionnaires in 2003 and 2012 to the present.

Table 2:

Comparison of OSSN management preferences from 2003 to 2022

	Percent of Survey Participants			P-Value (2022 vs. 2003)	P-Value (2022 vs. 2012)
	2003	2012	2022
First-line Topical Immuno/Chemotherapy
5FU	5	3	23	<0.0001	<0.0001
MMC	76	37	19	<0.0001	0.0007
IFN	14	56	55	<0.0001	0.9
Surgical Excision Alone
<2 mm	66	51	26	<0.0001	<0.0001
2-8 mm	64	25	12	<0.0001	0.004
>8mm	47	18	5	<0.0001	0.0001
Surgical Excision with Adjunctive Therapy
< 2 mm	7	20	16	0.01	0.4
2-8 mm	27	50	45	0.007	0.4
>8 mm	45	59	55	0.1	0.4
Medical therapy alone
<2 mm	12	25	54	<0.0001	<0.0001
2-8 mm	9	25	44	<0.0001	0.002
>8 mm	7	26	41	<0.0001	0.02
Diagnostic Biopsy
Never	11	9	31	<0.0001	0.0001
Sometimes	26	40	50	0.0002	0.3
Always	51	51	24	<0.0001	0.0001

OSSN- Ocular Surface Squamous Neoplasia; 5FU- 5-fluorouracil; IFN- interferon; MMC- mitomycin C

Use and concentrations of specific topical agents were assessed only in the present survey. Regarding IFN, 90% of respondents used IFNα2b, 11% used IFNα2a, and 1% used pegylated interferon. For MMC, 0.02% was the most commonly used strength (57%); the most common treatment regimen consisted of MMC 0.02% administered four times daily for one week followed a two-week holiday (33%). Regarding 5FU, the most commonly utilized treatment regimen consisted of 5FU 1% four times daily for one week followed by a three-week holiday (57%).

Regarding cost, 86% considered IFN therapy to be expensive, compared to MMC (19%) and 5FU (19%). Participants were then asked to rank factors other than efficacy that influenced their drug of choice from 1 (most important) through 4 (least important). An overall score was then calculated based on the frequency of assigned ranks with higher scores considered to be more influential on drug selection. The most important factor in selecting an agent was toxicity at 3.07, followed by drug accessibility at 3.06, drug cost at 2.35, and the lowest was the dosing regimen at 1.59.

From 2003 to 2022, the frequency of using medical therapy alone significantly (p<0.0001) increased for lesions of all sizes. Specifically, for lesions less than 2 millimeters, topical monotherapy use increased from 12% to 54%, for lesions between 2-8 millimeters 9% to 44%, and in lesions measuring greater than 8 millimeters the change was a 7% to 41%. The frequency of performing excision alone significantly (p<0.0001) decreased from 66% to 26% for lesions less than 2 millimeters, 64% to 12% for lesions between 2 and 8 millimeters, and 47% to 5% for lesions greater than 8 millimeters from 2003 to 2022.

OSSN treatment approaches were examined by lesion size in our 2022 survey and compared to the prior studies. Among OSSN lesions measuring < 2 millimeters (Figure 1), the majority (54%) of individuals reported medical therapy alone as the preferred approach, with IFN therapy as the most popular (30%) monotherapy. Twenty-six percent of participants used excision alone in this size group.

Figure 1:

Management preferences for OSSN lesions that measure less than 2 millimeters.

* Represents a statistically significant difference (p<0.05) from 2012 to 2022

** Represents a statistically significant difference (p<0.05) from 2003 to 2022†

† Column represents a summed total of the medical therapy therapy options

With lesions ranging from 2-8 millimeters (Figure 2), excision with postoperative medical therapy (45%) was similar to those preferring medical therapy alone (44%). When evaluating only the medically treated lesions in this size group, 50% of clinicians used topical IFN, 30% used topical 5FU, 11% used topical MMC, and 10% combined topical and injected IFN. Surgical excision alone was the least utilized modality at 12%.

Figure 2:

Management preferences for OSSN lesions that measure between 2 and 8 millimeters.

* Represents a statistically significant difference (p<0.05) from 2012 to 2022

** Represents a statistically significant difference (p<0.05) from 2003 to 2022†

† Column represents a summed total of the medical therapy therapy options

Despite a significant (p<0.0001) increase in medical therapy from 7% to 41%, from 2003 to 2022 in lesions > 8 millimeters, excision with postoperative medical therapy remained the most utilized treatment approach at 55%, followed by primary medical therapy alone at 41% (18% topical IFN, 10% topical 5FU, 6% combined topical and injected IFN, 5% topical MMC, 1% IFN injection therapy) (Figure 3).

Figure 3:

Management preferences for OSSN lesions that measure greater than 8 millimeters

* Represents a statistically significant difference (p<0.05) from 2012 to 2022

** Represents a statistically significant difference (p<0.05) from 2003 to 2022†

† Column represents a summed total of the medical therapy therapy options

Surgical technique and postoperative therapy for OSSN were variable amongst the participants. We found that the most commonly employed margin size was 2 millimeters (46%). This is consistent with the mean margin size reported in 2003 (2.3 ± 0.84 mm), and mean margin sizes of 2-4 mm reported by 85% of participants in 2012. Intraoperative adjunctive therapy included cryotherapy to the conjunctival margins (79%), absolute alcohol therapy (45%), cryotherapy to the base of the excised lesion (29%), cryotherapy to the limbal margins (34%), and MMC intraoperatively (18%).

When performing surgical excision regardless of size, 61% of respondents used postoperative topical adjunctive medical therapy in eyes with positive margins, and 31% of respondents used postoperative topical adjunctive medical therapy in all patients. For tumors with negative margins post-excision, 16% of respondents use postoperative topical therapy for tumors greater than 8 millimeters in size, 5% of respondents use postoperative topical therapy for tumors less than 8 millimeters in size, and 7% of respondents do not use postoperative topical therapy. Seventy-eight percent of participants in 2003 reported applying cryotherapy to the conjunctival margins and 62% applied cryotherapy to the base of the excision site or limbus, which do not represent statistically significant differences when compared to 2022.

Discussion

In the present study, we examined trends in the management of OSSN. Overall, we found a significant (p<0.0001) increase in the use of topical immuno/chemotherapy and a reduction (p<0.0001) in the frequency of excision without adjunctive postoperative immuno/chemotherapy among OSSN lesions of all sizes over the past two decades.³⁴ The average rate of excision alone for all lesion sizes was 57% in 2003 versus 16% in 2022. This shift is likely due to the high recurrence rate noted with surgical excision alone, as high as 56% in cases with positive and 33% in cases with negative margins.³⁵ While this survey demonstrated that 2 mm margins were most commonly utilized, various outcomes are seen with margins between 2-4 mm.³⁶ Shields et al. originally described the use of 4 mm margins to minimize the likelihood of positive surgical margins.³⁷ However, a margin size of 2 mm with adjunctive cryotherapy was also found to not be associated with high rates of positive postoperative biopsy results.³⁸

It is thus not surprising that a combined surgical excision with adjuvant therapy, including intraoperative cryotherapy and postoperative topical therapy, are commonly utilized. For example, in one study, tumors treated with intraoperative cryotherapy had lower recurrence rates at 5 years (31% versus. 16%) compared to tumors where cryotherapy was not used.³⁹ In a similar manner, postoperative IFN therapy in individuals with positive tumor margins decreased recurrence rates to a similar level of those with excised OSSN and negative margins.^{39, 40} Using a literature-based decision analysis, Siedlecki et al. found that surgical excision with postoperative IFN therapy was the most effective management strategy for cases of OSSN with positive margins.⁴¹

Looking at specific topical agents, overall IFN remained the favored medical treatment at 14% in 2003, 56% in 2012, and 55% in our present study. There was a notable increase in the use of 5FU, which was only used in 3% of participants in 2012 and increased (p<0.0001) to 23% in our present study. For those using topical monotherapy, the frequency of MMC use decreased (p=0.0007), from 37% in 2012 to 19% in 2022. Factors that seem to drive these choices were efficacy and side effect profile, as IFN was the most popular choice despite having the lowest accessibility (62%) and highest cost in the U.S. (~$800 a month for self-pay patients).^{42, 43}

It is also interesting that an increased proportion of participants responded that they “never” obtained a biopsy before initiating medical therapy when compared to the survey in 2003 (9% vs. 31%). These results may reflect the increased reliance upon high resolution anterior segment optical coherence tomography (AS-OCT) in serving as an adjunct to establish a diagnosis of OSSN. AS-OCT has demonstrated a sensitivity and specificity as high as 100% in differentiating OSSN from other ocular surface lesions and shown efficacy in screening for disease recurrence. ^{44, 45} In any lesion with atypical features, or unresponsive to medical therapy, tissue evaluation is needed to avoid missing a masquerading lesion. Looking forward, we suspect that increased utilization of deep learning methods as diagnostic tools in ophthalmology⁴⁶ will lead to future applications of this technology to help identify and monitor treatment of OSSN. Finally, other hypotheses for the lower rates of biopsy can be considered, as AS-OCT has not been universally adopted worldwide. These may include clinicians’ increased reliance on clinical exam alone to establish a diagnosis of OSSN and willingness to initiate medical therapy first, given newer literature demonstrating high success rates and low complication rates.

As prior studies grouped lesions by size, we also compared responses based on lesion size. The use of medical management in 2022 was reported as 54%, 44%, and 41% across the < 2 mm, 2-8 mm, and > 8 lesion size groups respectively. These values were increased as compared to 2003 (p<0.0001).

In our current survey results, primary medical therapy use was preferred management for lesions < 2 millimeters. This represents a change from the 2012 survey, where tumor excision alone (without postoperative topical therapy) was preferred for OSSN lesions < 2 millimeters. For lesions 3-8mm in size, the 2022 survey showed a decrease in the use of excision alone, and equal use of topical medical therapy alone or excision with topical therapy.³⁴

For lesions greater than 8 mm, the majority of respondents in 2022 preferred surgery with postoperative medical therapy. This is interesting because this approach goes against our personal treatment preferences. With increasing tumor size, surgical excision often requires removal of more limbal stem cells, depending on tumor location. We have found that limbal stem cell deficiency becomes more frequent when > 25% of the limbus is destroyed with surgery. Our experience and previous studies have demonstrated complete tumor resolution in cases of large tumors with extensive limbal involvement after treatment with primary medical therapy alone, with minimal side effects.^47-49 As such, we prefer initiating larger lesions, if possible, with topical agents. In many cases this will result in tumor resolution, but at a minimum, the use of neoadjuvant and chemo/immune-reduction agents may allow for tumor shrinkage and thus less extensive surgery.

Our study has several limitations. The majority of the surveyed participants considered themselves to be cornea specialists and thus the sample may not represent all clinicians who treat OSSN. In addition, we do not have information on the use of adjuvant diagnostic tools (e.g., AS-OCT) that would help interpret the study results. Furthermore, there are other complexities, such as clinical features of tumor type (dysplasia vs SCC) which may drive therapeutic decisions but were not included in our current survey. This survey was administered while IFN was still available in the US. Since 2022, IFN is currently not produced in the United States resulting in decreased accessibility for patient care. It is still readily available in certain countries, such as India. Due to this limited accessibility in the US, we expect 5FU to be the first-line replacement option for medical therapy until interferon is again available. We have previously shown that 5FU has demonstrated comparable rates of resolution in the treatment of OSSN and lower cost, but carries a more significant side effect profile.^{12, 43} Finally, the online link to access the survey was provided during presentations at academic meetings, thus an accurate response rate for eligible participants is unable to be calculated.

Despite these limitations, the present study demonstrates a continued paradigm shift in the management of OSSN. The use of primary medical therapy as the initial approach has increased significantly, in lesions of all sizes. Similarly, there was a decrease in the utilization of surgical excision alone, without the use of postoperative medical therapy. In addition, topical interferon remains the preferred topical agent, and 5FU has grown significantly in popularity as primary medical therapy.

Supplementary Material

Supplemental Data File (.doc, .tif, pdf, etc.)

Appendix 1: Questions included in the 2022 survey