Managing ocular surface neoplasia without biopsy: The end of pathology as we know it?

Abstract

Ocular surface squamous neoplasia encompasses a wide variety of lesions with very different biological behaviours and prognoses. This umbrella term incorporates benign papillomas, intraepithelial neoplasias, and invasive carcinoma, including those with high risk histology (adenosquamous, spindle cell/sarcomatoid squamous cell carcinoma). For this reason, the eighth edition of the American Joint Committee on Cancer recommends limiting the term “OSSN” to clinical usage only. In this issue, Polski et al describe a retrospective case series of 41 eyes of ocular surface neoplasia that compares biopsied tumours with known depth invasion with those staged clinically without biopsies. The authors conclude that, because there were no significant differences in outcomes between invasive and non-invasive tumours, treatment (with topical therapy) can be undertaken without histological assessment and with clinical staging alone. Certainly provocative, this conclusion requires considerable further validation and research in properly designed prospective multi-institutional studies. The small sample size and smaller number of recurrences make it impossible to make meaningful clinical recommendations. In addition, there are several other weaknesses with the data presented. First, follow-up period is not defined. Patients with in situ vs invasive disease are not separately analysed; the term OSSN is used both histologically and clinically. All cases in the series had multiple structure involvement, yet it is not clear whether they represented recurrences (in which a previous biopsy was performed) or primary lesions. Regarding the value of optical coherence tomography (OCT) in the clinical staging of OSSN, data presented in the current study are also limited. However, this is a promising area warranting further investigation, with a larger prospective study examining the correlation between specific OCT parameters and histopathology. While all data on this much neglected group of diseases are welcome, we caution that there are important principles of current best practice cancer care that need to be considered. Australian and international cancer guidelines increasingly promote a multidisciplinary model in which management decisions are rarely made without tissue diagnosis. There are sound reasons for this. First, amelanotic or minimally pigmented conjunctival melanoma, which is clinically indistinguishable from OSSN and estimated to occur in up to 30% of cases in some series, requires a tissue diagnosis.3–6 A number of studies have highlighted the very poor outcomes in conjunctival melanoma patients in which the diagnosis is unsuspected and the treatment is unplanned.6–8 Second, some ocular surface carcinomas are associated with more aggressive behaviour (eg, perineural invasion, multifocal disease and orbital invasion) and poorer prognosis; these include poorly differentiated and spindle cell squamous cell carcinoma, adenosquamous carcinoma, Merkel cell carcinoma and sebaceous carcinoma. Such cases perform poorly in the experience of ourselves and others when the diagnosis is unsuspected, and a recurrence is eventually biopsied.6–8 In addition, there is the risk of overtreatment, when benign conditions mimicking a neoplastic process are subjected to ongoing topical drug treatment. While a conservative approach to treatment is always a goal, topical therapy, like surgery, is not without morbidity and cost to the patient. With rare exceptions such as retinoblastoma (where biopsy is potentially hazardous), it is difficult to justify bypassing biopsy in the era of precision and personalized cancer medicine, where management is increasingly based on specific biomolecular tumour pathways that require tissue sampling. A case in point is the trend we are currently seeing for the management of uveal melanoma, where tissue sampling is increasingly being undertaken to guide the management of these patients. While this does not apply to OSSN now, it may be that strategies in the future will require more, not less, intimate knowledge of tumour genomics and immune microenvironment, as has occurred for many other cancers. The understanding of the natural history of ocular surface neoplasia is evolving, and this article can be seen as a call for more international collaboration on the diagnosis, classification and outcomes of OSSN. As OSSN lesions are relatively rare, large multi-site collaborative registries are one approach to collect sufficient data for evidence-based changes to management in the absence of prospective trials. A similar approach is currently being undertaken in ocular naevi and melanoma. It is overdue that this rigour be incorporated into DOI: 10.1111/ceo.13475