Background/aims Fine needle aspiration biopsy has become a standard technique to obtain material from primary uveal melanoma for diagnosis, prognostication and research. Cytopathology requires a significantly greater tissue yield for a positive diagnosis compared to other molecular tests. The purpose of this report was to compare sample retrieval for cytopathology between trans-scleral and vitrectomy-assisted transvitreal biopsy in the same tumour in patients who underwent iodine-125 brachytherapy.
Methods All clinical records of patients who underwent iodine-125 brachytherapy for the treatment of uveal melanoma in conjunction with pars plana vitrectomy from whom trans-scleral and vitrectomy-assisted transvitreal fine needle biopsies were obtained, were reviewed. Baseline patient and tumour characteristics, biopsy sample yield for cytopathology, and surgical outcomes were reviewed.
Results Thirty-eight patients with a median follow-up time of 2 months were included. Overall, 27 of 38 (71%) of transvitreal biopsies, and 25 of 38 (66%) of trans-scleral biopsies yielded sufficient material for cytopathologic analysis (p=0.62). For transvitreal and trans-scleral biopsies, larger tumours were associated with significantly higher cytopathologic yield, but tumour location did not significantly affect cytopathologic yield.
Conclusions Trans-scleral and vitrectomy-assisted transvitreal fine needle aspiration biopsy resulted in similar tissue yield for cytopathologic analysis. As molecular prognostication becomes the standard of care for obtaining patient information and directing clinical management, further investigation of biopsy techniques is warranted.
- Treatment Surgery
Statistics from Altmetric.com
Uveal melanoma is the most common primary intraocular malignancy in adults, with an annual incidence of 5.1 persons per million per year in North America.1 Metastasis occurs in up to 50% of patients, with death generally occurring within 1 year of detection of metastasis.2 To date, there are no successful therapies for metastatic uveal melanoma.3 Fine needle aspiration biopsy (FNAB) of tumours at the time of local therapy has recently emerged as an important tool in the management and study of these patients.4 In addition to cytopathologic diagnostic confirmation, FNAB may provide molecular prognostication to identify patients at high risk for metastatic disease. Knowledge of metastatic risk may allow for individualised patient-centred therapy through altering the schedule for systemic surveillance as well as eligibility for clinical trials of adjuvant therapy.4 Finally, FNAB provides material for cultivating cell lines that may be used to better understand the molecular biology of metastasis, which may, in turn, lead to targeted therapies for metastatic disease.5
There has been controversy surrounding FNAB at the time of local treatment of uveal melanoma, as some have speculated it may increase the risk of orbital dissemination, local recurrence, and/or metastatic disease.6 ,7 However, a recent report from our centre with up to 6 years of follow-up of 170 patients with uveal melanoma who underwent 30-gauge trans-scleral FNAB at the time of brachytherapy found no case of orbital dissemination or local recurrence, and there was no increased metastatic risk.8
There is no standard technique used to perform uveal melanoma FNAB. Authors have reported successful FNAB by transvitreal and trans-scleral approaches,8–10 and it is not known whether a transvitreal or trans-scleral approach is superior for obtaining tissue. However, there have been no reports in the literature comparing the yield of transvitreal and trans-scleral approaches in the same tumour, or tumours matched by size and location.
We report the biopsy yield for cytopathologic sample retrieval in a consecutive series of 38 uveal melanomas that underwent vitrectomy-assisted transvitreal and trans-scleral FNAB at the time of brachytherapy.
Materials and methods
The study was performed in accordance with the US Health Insurance Portability and Accountability Act (HIPAA) of 1996, and was approved by the Office of the Human Research Protection Programme (Institutional Review Board) of the University of California, Los Angeles. Charts of all patients with a clinical diagnosis of uveal melanoma who underwent intraoperative vitrectomy-assisted transvitreal and trans-scleral FNAB at the time of plaque brachytherapy at Stein Eye Institute, UCLA, between 1 September 2011 and 8 July 2013, were reviewed. Two patients were excluded because they did not consent to participate in clinical research. Additionally, 13 patients were excluded because tumour material was sent for cytogenetic analysis rather than cytopathologic analysis. Finally, one patient was excluded because the tumour material source (transvitreal or trans-scleral biopsy) could not be determined during cytopathologic analysis.
Diagnosis of uveal melanoma was established by comprehensive ophthalmic examination, ultrasonography, and ﬂuorescein angiography. Systemic evaluation by an oncologist or internal medicine specialist revealed no evidence of melanoma, metastasis, or other active primary cancer, and psychologic support was offered by a clinical psychologist or social worker with expertise in uveal melanoma.
Trans-scleral and transvitreal biopsies were performed at the time of plaque brachytherapy for the purpose of obtaining sufficient tissue for cytopathologic and cytogenetic clinical testing, and for the research purposes of tissue culture and nucleotide analyses. All patients underwent concomitant pars plana vitrectomy with silicone oil fill for radiation attenuation, and in some cases, for serous retinal detachment.11 ,12 Iodine-125 plaque surgery with FNAB and vitrectomy was performed by a single surgeon (TAM). The details of FNAB and radioactive plaque placement have been described.9 ,13–15 In brief, trans-scleral FNAB was performed with a short 30-gauge needle via a tangential approach. Typically, six to eight trans-scleral passes were performed. No adjunctive tissue adhesives were used. The plaque was sutured in place with optimal plaque position achieved with intraoperative ultrasonography. Following a complete 23-gauge vitrectomy including detachment of the posterior hyaloid and peripheral vitreous base shave, the vitrectomy-assisted transvitreal biopsy was performed using a 1.5 inch 27 gauge needle, with one or two passes. Air-fluid exchange preceded silicone oil 1000 centistokes fill, and the sclerotomy sites were sutured in all cases.
During trans-scleral and transvitreal biopsies, an assistant provided aspiration via a 10-cm3 syringe connected to the needle by tubing. The biopsy material was immediately smeared onto glass slides, fixed in ethanol, and stained with haematoxylin-eosin. Indeterminate samples were later tested by immunohistochemistry for the human melanoma black-45 antibody.
Cytology of the biopsies was evaluated by an ophthalmic pathologist, who separately reported the findings from trans-scleral and transvitreal biopsies. We chose to analyse yield of cytopathology rather than cytogenetics because of the possible ambiguity of certain cytogenetic results. A disomy 3 result in fluorescence in situ hybridisation (FISH) analysis could represent either disomy 3 tumour material or normal tissue that has inadvertently been biopsied.
Patient charts were retrospectively reviewed and the following data were collected: demographic information; tumour parameters; date of brachytherapy surgery; trans-scleral and transvitreal FNAB cytopathology; complications related to FNAB including localised or diffuse vitreous haemorrhage and rhegmatogenous retinal detachment; local treatment failure; uveal melanoma metastatic disease and/or tumour-related death; and last date of follow-up. Length of follow-up was determined by calculating the number of days between date of brachytherapy surgery and last date of follow-up. Patient data were tabulated using Excel (Microsoft Ofﬁce Excel 2003; Microsoft, Redmond, Washington, USA) and FileMaker Pro (FileMaker Pro 12.0v4; FileMaker, Santa Clara, California, USA).
The χ2 test was used to compare categorical variables, and two-tailed Student t test was used to compare quantitative variables. When a value less than 5 was present in a contingency table, Fisher's exact test was used instead of the χ2 test. A p value less than 0.05 was considered statistically significant.
Fifty-four patients underwent vitrectomy-assisted transvitreal and trans-scleral FNAB at the time of brachytherapy between 1 September 2011 and 8 July 2013. Two patients declined participation in clinical research and were excluded from analysis. Additionally, 15 patients were excluded because tumour material was sent for cytogenetic rather than cytopathologic analysis, and one patient was excluded because the tumour material source (transvitreal or trans-scleral biopsy) could not be determined at the time of cytopathologic analysis. The remaining 38 patients were included in this study. The characteristics of these patients and their tumours are shown in table 1. The mean age at diagnosis was 61.2 years (range 30–83 years); 22 patients were male, and 16 patients were female; 15 tumours were located in the right eye, and 23 tumours were located in the left eye. The mean tumour height was 4.8 mm (range 1.7–13.3 mm), and the mean tumour largest basal diameter was 10.7 mm (range 5.5–15.9 mm). Two tumours (5%), 15 tumours (40%), 10 tumours (26%), and 11 tumours (29%) were located in the ciliary body, peripheral choroid, equatorial region, and posterior pole, respectively. The median follow-up time was 2 months.
The complications and outcomes of the 38 patients included in the study are shown in table 2. Fifteen eyes (39%) were noted to have localised vitreous haemorrhage postoperatively. Because tumours underwent transvitreal and trans-scleral biopsies at the time of surgery, we could not determine which type of biopsy caused the haemorrhage. No eyes developed diffuse or non-clearing vitreous haemorrhage or rhegmatogenous retinal detachment as a result of FNAB. During the follow-up interval, no patients developed local recurrence or metastatic disease.
Overall, 71% of transvitreal biopsies and 66% of trans-scleral biopsies yielded sufficient material for cytopathologic analysis (table 3). This difference was not statistically significant (p=0.62). Transvitreal and trans-scleral approaches had statistically significantly higher yields in tumours that were larger in height (table 4). Among transvitreal biospies, the yield was 44%, 92%, and 90% in tumours with heights of less than 3.0 mm, 3.0 to 5.0 mm, and greater than 5.0 mm, respectively (p=0.03). Among trans-scleral biopsies, the yield was 38%, 75%, and 100% in tumours with heights of less than 3.0 mm, 3.0 to 5.0 mm, and greater than 5.0 mm, respectively (p=0.015). As shown in table 5, there was no difference in yield between transvitreal and trans-scleral approaches among tumours with heights of less than 3.0 mm (p=0.72), 3.0 to 5.0 mm (p=0.59), or greater than 5.0 mm (p=1.0).
Tumour largest basal diameter (LBD) significantly influenced tumour yields for transvitreal and trans-scleral approaches (figure 1). Among transvitreal biopsies, the mean tumour LBD in biopsies that yielded sufficient material for cytopathologic analysis was 11.4 mm, compared to 9.1 mm in insufficient biopsies (p=0.02). Among trans-scleral biopsies, the mean tumour LBD in sufficient biopsies was 11.6 mm, compared to 9.1 mm in insufficient biopsies (p=0.007).
Tumour location did not significantly influence biopsy yield for either transvitreal or trans-scleral approaches, although there was a trend toward higher yield in ciliary body and equatorial tumours for both approaches (table 6). Among transvitreal biopsies, 100%, 60%, 100%, and 54.5% of tumours located in the ciliary body, peripheral choroid, equatorial region, and posterior pole, respectively, yielded sufficient material for cytopathologic analysis (p=0.26). Among trans-scleral biopsies, 100%, 66.7%, 100%, and 45.5% of tumours located in the ciliary body, peripheral choroid, equatorial region, and posterior pole, respectively, yielded sufficient material for cytopathologic analysis (p=0.63). As shown in table 7, there was no difference in yield between transvitreal and trans-scleral approaches in tumours located in the ciliary body (p=1.0), peripheral choroid (p=0.7), equatorial region (p=1.0), or posterior pole (p=1.0).
We present the first data comparing the yield of vitrectomy-assisted transvitreal to trans-scleral fine needle aspiration biopsies obtained from the same uveal melanoma. Cytopathology requires a significantly greater amount of material at biopsy for a positive diagnosis compared to, for example, gene expression profiling. Overall, we report that 71% of transvitreal biopsies and 66% of trans-scleral biopsies were sufficient for cytopathologic analysis, which is consistent with previous reports.8 ,10 ,16 ,17 There was no significant difference in yield between the two approaches. Moreover, the yield was significantly improved in tumours that were larger by height and largest basal diameter, and there was no significant difference in yield based on tumour location. Transvitreal and trans-scleral approaches did not differ significantly in tumours of differing sizes or locations.
Our finding that larger tumour size is associated with higher FNAB yield is consistent with prior investigations.8 ,17 Our recent study of 170 patients with uveal melanoma who underwent intraoperative trans-scleral FNAB with a 30-gauge needle found that sufficient material was obtained for fluorescent in situ hybridisation (FISH) analysis in 53%, 68% and 91% of patients with tumours less than 3.0 mm, 3.0–5.0 mm, and greater than 5.0 mm in height, respectively.8 Similarly, Cohen et al reported a significant effect of lesion height on sufficiency of FNAB obtained transvitreally for cytologic examination of suspicious choroidal masses.17 In their study, diagnostic material was obtained in 40% of lesions less than 1.99 mm thick, 90% of lesions between 2.00 and 4.00 mm thick, and 98% of lesions greater than 4.00 mm thick. We hypothesise that this is due to greater accuracy of localising the needle within the tumour when the target size is larger. In a study of six trans-scleral and 16 transvitreal cases of 30-gauge FNAB in enucleated eyes with uveal melanoma, Glasgow et al found that the single case in which biopsy was inadequate for cytopathologic analysis occurred because the needle puncture site was 1 mm away from the tumour.18 Therefore, inadequate yield may be due to inaccurate needle localisation. For this reason, we recommend performing multiple needle passes in different tumour locations and verbal feedback with the assistant collecting biopsy material in order to maximise biopsy yield.
In addition to tumour height, we found tumour largest basal diameter to be significantly associated with biopsy yield. Again, this may be due to improved accuracy of needle localisation in a larger tumour.
We found no significant effect of tumour location on sufficiency of FNAB material for cytopathologic diagnosis. However, this may be due to the small number of patients in each group. For trans-scleral and transvitreal approaches, yield was lower when sampling tumours in the peripheral choroid and posterior pole compared to the ciliary body and equatorial region. This is consistent with the findings of Midena et al, who reported that although trans-scleral FNAB was sufficient for FISH analysis in 91% of ciliary body tumours and 76.8% of choroidal tumours, there was no significant effect of tumour location on FNAB yield.16
In the current analysis, there was no significant difference in the adequacy of vitrectomy-assisted transvitreal and trans-scleral biopsy sample for cytopathologic analysis, even when tumours of varying sizes and locations were considered. By contrast, Shields et al compared the yield of transvitreal and trans-scleral biopsies for DNA amplification and microsatellite assay.10 Transvitreal biopsies were performed on postequatorial tumours, and trans-scleral biopsies were performed on pre-equatorial tumours. They reported a higher yield for DNA-based testing in transvitreal biopsies (97%) compared to trans-scleral biopsies (75%), and hypothesised that this may have been due to greater accuracy from visualising the needle tip entering the tumour apex using indirect ophthalmoscopy with the transvitreal technique. Alternatively, they hypothesised that the smaller needle gauge used with the trans-scleral technique (30-gauge vs 27-gauge) may have contributed to the lower yield. However, because they sampled different tumours in different locations with the two techniques, and did not report on the relative sizes of the tumours sampled, the yields of the two methods are not directly comparable. Furthermore, it must be noted that tissue yield was reported for DNA amplification and microsatellite assay, rather than cytopathology, the latter generally requiring more tissue.
In our report of 38 eyes that underwent trans-scleral and vitrectomy-assisted transvitreal FNAB, 15 (39%) developed localised vitreous haemorrhage, which was self-limited in all cases. No patient developed diffuse or non-clearing vitreous haemorrhage, rhegmatogenous retinal detachment, local recurrence, or metastasis. This is consistent with previous reports of the safety of intraoperative fine needle aspiration biopsy.8 ,10 In our recent study of 170 patients with uveal melanoma who underwent 30-gauge trans-scleral FNAB at the time of brachytherapy, there were no cases of orbital dissemination or local treatment failure. There was no difference in the rate of metastasis compared to a group of patients with similarly sized tumours in the Collaborative Ocular Melanoma Study.8
While there are currently no successful treatments for metastatic uveal melanoma, the importance of obtaining prognostic information is highlighted by Beran et al, who showed that 97% of patients wish to know their prognosis for future planning.13 Furthermore, a poor prognostic biopsy result was not correlated with psychological morbidity.13 Moreover, many oncologists may choose to increase the frequency of systemic surveillance in patients with a poor prognostic biopsy result, and these patients may elect to enrol in clinical trials of adjuvant therapy for premetastatic disease.4 Clinical trials may require a molecular prognostic result for eligibility. As treatment options become available, material from the primary tumour may become more important in guiding individualised therapy. From a research standpoint, FNAB of uveal melanomas provides material for cell lines on which experiments may be conducted to help better understand the biology of metastasis. This, in turn, may guide future targeted therapies for metastatic uveal melanoma, which is currently almost uniformly fatal.5
The strengths of this study include a single centre, single examiner and single surgeon with consistent surgical technique, as well as an experienced ophthalmic pathologist. Limitations of our study include a relatively small number of patients with a short follow-up interval, making it difficult to assess rare and long-term complications. Additionally, because our study evaluated cytopathology, the results may not be directly applicable to DNA-based and RNA-based molecular testing. Moreover, biopsies were performed trans-sclerally with a 30-gauge needle and transvitreally with a 27-gauge needle, and yield may differ in biopsies performed with different size needle bores or different instrumentation, such as with a vitreous cutter. Finally, transvitreal biopsy was performed only at the time of pars plana vitrectomy, wherein a complete vitrectomy was performed prior to silicone oil fill for attenuation of iodin-125. Other clinicians perform transvitreal biopsy with visualisation using the indirect ophthalmoscope, which produces an inverted image, and biopsy yield may therefore differ when transvitreal biopsy is performed by this method.
In summary, we report no difference in yield between 27-gauge transvitreal and 30-gauge trans-scleral fine needle aspiration biopsies obtained at the time of pars plana vitrectomy and brachytherapy for cytopathologic diagnosis in 38 patients with uveal melanoma. Biopsies of larger tumours were more likely to be sufficient for cytopathologic analysis regardless of technique used, and there was no effect of tumour location on sample yield. As molecular prognostication becomes the standard of care for obtaining patient information and directing clinical management, further investigation of biopsy techniques is warranted.
Contributors Both authors contributed substantially to the conception and design, acquisition of data, and analysis and interpretation of data. Both authors drafted and revised the manuscript critically for intellectual content and approved of the final published version.
Funding Supported by an unrestricted grant from Research to Prevent Blindness and the George E and Ruth Moss Trust.
Competing interests None.
Ethics approval Office of the Human Research Protection Program (Institutional Review Board) of the University of California, Los Angeles.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.