January 2014
Volume 55, Issue 1
Free
Multidisciplinary Ophthalmic Imaging  |   January 2014
Relationship Between Rate of Posterior Uveal Melanoma Flattening Following Plaque Radiotherapy and Gene Expression Profile Class of Tumor Cells
Author Notes
  • Department of Ophthalmology, College of Medicine, University of Cincinnati, Cincinnati, Ohio 
  • Correspondence: Zélia M. Corrêa, Medical Arts Building, Suite 1500, Cincinnati, OH 45219; correazm@uc.edu
Investigative Ophthalmology & Visual Science January 2014, Vol.55, 556-559. doi:10.1167/iovs.13-13381
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Zélia M. Corrêa, James J. Augsburger; Relationship Between Rate of Posterior Uveal Melanoma Flattening Following Plaque Radiotherapy and Gene Expression Profile Class of Tumor Cells. Invest. Ophthalmol. Vis. Sci. 2014;55(1):556-559. doi: 10.1167/iovs.13-13381.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose.: To evaluate the relationship between rate of flattening of posterior uveal melanomas (PUMs) over the first 6 months following I-125 plaque radiotherapy and gene expression profile (GEP) class of tumor cells obtained by fine needle aspiration biopsy (FNAB) prior to treatment.

Methods.: Retrospective analysis of relationship between GEP of PUM cells obtained by FNAB at or prior to treatment and rate of tumor flattening following I-125 plaque radiotherapy. Impact of initial tumor thickness was minimized by pairing cases so baseline tumor thickness in subgroups was matched to within ± 0.5 mm. Paired t-testing compared mean tumor thickness in GEP subgroups at 3- and 6-months post treatment assessments.

Results.: Our initial group consisted of 269 patients. Seventy-seven tumors (28.6%) were GEP class 2. Twenty-seven of these were treated by I-125 plaque radiotherapy post-FNAB and returned for post treatment evaluations at 3 and 6 months. A matched GEP class 1 tumor was identified for 25 class 2 cases. Matched tumor pairs ranged in thickness from 2.5 to 11.5 mm at baseline. Mean tumor thickness at baseline in the GEP 1 subgroup was 5.8 and 5.9 mm in the GEP 2 subgroup. Three-months post plaque, mean tumor thickness was 4.5 mm in class 1 cases and 4.6 mm in class 2 cases (paired t = 0.31, P = 0.76). The 6-month post-plaque, mean tumor thickness was 4.0 mm in each subgroup (paired t = 0.25, P = 0.81).

Conclusions.: Our study showed a lack of association between the GEP class and the rate of flattening of posterior uveal melanomas following I-125 plaque radiotherapy of PUMs.

Introduction
In 1987, Augsburger and coworkers 1 first reported that faster rate of clinical flattening of radioactive plaque–treated posterior uveal melanomas during the first 6-months post treatment was an independent, significant prognostic factor for metastasis and metastatic death not explained by the combination of patient age at treatment, largest linear dimension of the primary intraocular tumor, and intraocular location of the intraocular tumor. In 1989, Glynn and coworkers 2 reported essentially the same finding in patients with posterior uveal melanomas treated by proton beam irradiation. In 1990, Guthoff and coworkers 3 confirmed the prognostic significance of faster clinical flattening of posterior uveal melanomas treated by Ru-106 plaque brachytherapy. In 2004, Kaiserman and coworkers 4 revisited the previously reported impact of rate of clinical flattening of posterior uveal melanomas following plaque radiotherapy. 
During the past decade, gene expression profile testing and classification of posterior uveal melanomas has been shown to be a stronger prognostic factor for metastasis and metastatic death than any previously identified individual clinical variable or combination of clinical variables. 5 The reports of the impact of rate of tumor flattening mentioned in the preceding paragraph were all performed prior to the development of the currently available prognostic gene expression profile (GEP) test (Decision Dx-UM; Castle Biosciences, Friendswood, TX). 6 Because of this, we do not know whether post-irradiation regression of posterior uveal melanomas is strongly correlated with GEP class of the tumor cells, and therefore without prognostic significance when GEP class is known or is independently prognostic of survival time in combination with GEP class. 
One group recently reported the relationship between the GEP class of tumor cells obtained by fine needle aspiration biopsy (FNAB) prior to primary tumor treatment and tumor regression following proton beam irradiation. 7 This group reported no difference in the rate of regression of GEP class 1 and class 2 tumors. To date, there have been no reports of the relationship between GEP class of tumor cells and rate of tumor regression following plaque radiotherapy. The purpose of this study was to determine whether the rate of post-irradiation flattening of posterior uveal melanomas differed significantly among tumors identified as GEP class 1 versus class 2 based on the analysis of tumor cells obtained by FNAB at the time of or shortly prior to I-125 plaque radiotherapy. 
Methods
We performed a retrospective analysis of the relationship between the GEP class of posterior uveal melanoma cells obtained by FNAB at, or shortly prior to, the start of treatment and rate of post-irradiation tumor flattening during the first 6-months post treatment in patients managed by I-125 plaque radiotherapy. To minimize the impact of initial tumor thickness on rate of tumor flattening, we employed a paired case approach in which tumor thickness and largest basal diameter (LBD) at baseline in class 1 and class 2 cases were matched as described ahead. Paired t-testing was used to compare the mean thickness of the tumors in the GEP groups at the 3- and 6-month post treatment visits. 
This research followed the tenets of the Declaration of Helsinki. All patients signed an informed consent after explanation of the nature and possible consequences of the FNAB. The FNAB for GEP testing was approved as a prospective protocol by the University of Cincinnati's institutional review board (UC-IRB). The follow-up and retrospective data retrieval was approved under the Ocular Tumors Database by the UC-IRB. 
Our study was planned to evaluate the comparative rates of flattening of primary posterior uveal melanomas following I-125 plaque radiotherapy that were classified by GEP testing of FNAB aspirates. To assure that we were comparing similar patients and tumors, we specified the following inclusion and exclusion criteria and matching criteria. 
In order to be included in this study a patient had to present (1) a primary uveal melanoma, (2) treated by I-125 plaque radiotherapy, (3) sampled by FNAB prior to treatment, and (4) evaluated post treatment at 3 and 6 months (±1 month) by ultrasound. Additionally, (5) FNAB aspirates submitted for both cytopathological analysis and GEP testing had to yield a result. 
Cases were excluded if (1) the tumor was limited to the iris, iris and ciliary body, or ciliary body (i.e., any tumor not involving the choroid exclusively or both the choroid and ciliary body), (2) the intraocular tumor had been previously treated, (3) the tumor had any atypical clinical features (i.e., any nonclassic melanocytic uveal tumor) at baseline, and (4) either the cytopathological or GEP test had failed. Failure of cytopathological analysis to confirm uveal melanoma included insufficient aspirates, nonmelanoma tumor, lost specimen, and technical errors. Failure of the GEP test included microarray technique unable to detect all three of the nonclassifying genes, 8 and low-confidence GEP classification of tumor cells (i.e., a discriminant score by the support vector machine [SVM] algorithm 9 between +0.1 and −0.1). 
Each patient with a GEP class 2 tumor who met the selection criteria specified above was matched with a patient with a GEP class 1 tumor who also met these selection criteria using the following matching criteria: 
  1.  
    Baseline LBD (by fundus mapping) within ±1 mm (all fundus mapping with estimation of LBD was performed by one examiner [JJA], whose accuracy of tumor basal dimension measurements has been reported 10 ); and
  2.  
    Baseline maximal tumor thickness (by ultrasonographic biometry) within ±0.5 mm (Each ultrasonographic biometry measurement was made by one of the authors [ZMC or JJA] who performed the ultrasonographic study).
The individual who performed the matching (JJA) was masked as to the regression status of the tumor or survival status of the patients. Patients from the GEP class 2 subgroup who could not be matched successfully with a patient from the GEP class 1 subgroup were excluded from our analysis dataset. Statistical analysis was performed by paired t-testing to compare the mean thickness of the tumors in the GEP subgroups at the 3- and 6-month post treatment assessments using the IBM SPSS software (V.22.0.0; Statistical Package for the Social Sciences, Armonk, NY). 
Results
Our base study group included patients recruited from September 2007 through March 2012 and consisted of 269 patients. Seventy-seven of these tumors (28.6%) were GEP class 2. Twenty-seven of these 77 patients were treated by I-125 plaque radiotherapy post-FNAB and returned for ultrasonographic measurement of tumor thickness at both 3- and 6-months post treatment. We identified a matching patient from the GEP class 1 tumor cases for 25 of the class 2 cases. The comparative baseline clinical features of the patients in the matched GEP subgroups are summarized in the Table. The tumors in the 25 pairs of patients ranged in LBD from 8.5 to 17.0 mm at baseline (mean 12.6 mm). The mean largest basal tumor diameter was 12.6 mm (SD = 2.6 mm) in the GEP class 1 group, and 12.5 mm (SD = 2.5 mm) in the GEP class 2 group at baseline. The tumors in the 25 matched pairs of patients ranged in thickness from 2.5 to 11.5 mm at baseline (mean 5.9 mm). The mean tumor thickness was 5.8 mm (SD = 2.2 mm) in the GEP class 1 group, and 5.9 mm (SD = 2.2 mm) in the GEP class 2 group at baseline. 
At 3-months post plaque, mean tumor thickness was 4.5 mm (SD = 1.6 mm) in the class 1 cases and 4.6 mm (SD = 1.8 mm) in the class 2 cases (paired t = 0.31, P = 0.76). At 6-months post plaque, mean tumor thickness was 4.0 mm (SD = 1.4 mm) in the class 1 cases and 4.0 mm (SD = 1.6 mm) in the class 2 cases (paired t = 0.25, P = 0.81). The comparative tumor regression in the two GEP class subgroups is shown graphically in Figure 1. Tumor flattening in the entire group is illustrated by a scatterplot graph in Figure 2. In terms of relative (percentage) thickness regression, at 3 months, the mean relative regression for the class 1 tumors was 20.9% and for class 2 tumors was 21.0%. At 6 months, the mean relative regression for the class 1 tumors was 28.6% and for the class 2 tumors it was 31.4%. 
Table
 
Comparison of Baseline Variables in GEP Class 1 and GEP Class 2 Patient Groups That Were Matched for Baseline LBD and Maximal Thickness of the Primary Posterior Uveal Melanoma
Table
 
Comparison of Baseline Variables in GEP Class 1 and GEP Class 2 Patient Groups That Were Matched for Baseline LBD and Maximal Thickness of the Primary Posterior Uveal Melanoma
Variable GEP Class 1 Categories GEP Class 2 Subgroup Chi-Subgroup df 2 P
Number (%) Number (%)
Age of patient, y
 ≤50 7 (28.0) 1 (4.0) 6.46 2 0.040
 >50 but ≤70 14 (56.0) 15 (60.0)
 >70 4 (16.0) 9 (36.0)
Sex of patient
 Male 12 (48.0) 9 (36.0) 0.74 1 0.39
 Female 13 (52.0) 16 (64.0)
LBD of tumor, mm
  ≤10 7 (28.0) 7 (28.0) 0.17 2 0.92
  >10 but ≤15 14 (56.0) 15 (60.0)
  >15 4 (16.0) 3 (12.0)
TH of tumor, mm
  ≤3.5 4 (16.0) 4 (16.0) 0.00 2 1.0
  >3.5 but ≤7 16 (64.0) 16 (64.0)
  >7 5 (20.0) 5 (20.0)
Intraocular tumor location
 Exclusively choroidal 19 (76.0) 19 (76.0) 0.00 1 1.0
 Involving ciliary body 6 (24.0) 6 (24.0)
Melanoma cell type
 Spindle 11 (44.0) 6 (24.0) 2.23 2 0.33
 Mixed 8 (32.0) 11 (44.0)
 Epithelioid 6 (24.0) 8 (32.0)
Figure 1
 
Graph showing mean tumor thickness ± SD of the mean for tumor thickness at baseline, 3- and 6-months post irradiation for the posterior uveal melanomas in the two patient groups.
Figure 1
 
Graph showing mean tumor thickness ± SD of the mean for tumor thickness at baseline, 3- and 6-months post irradiation for the posterior uveal melanomas in the two patient groups.
Figure 2
 
Scatterplot graph showing tumor regression pattern by measurement of tumor thickness at baseline, 3-, and 6-months post irradiation for the posterior uveal melanomas in all 50 tumors studied.
Figure 2
 
Scatterplot graph showing tumor regression pattern by measurement of tumor thickness at baseline, 3-, and 6-months post irradiation for the posterior uveal melanomas in all 50 tumors studied.
Discussion
Based on prior studies identifying that the faster the rate of clinical tumor flattening during the first 6-months post treatment after radiation treatment to be an independent significant prognostic factor for metastasis and metastatic death not explained by the combination of patient age at treatment, largest linear dimension of the primary intraocular tumor, and intraocular location of the intraocular tumor, 1 -4 we hypothesized that GEP class would be strongly associated with faster post–plaque tumor shrinkage. 
Marathe and coworkers 11 have recently reported on the response of choroidal melanomas with monosomy 3 versus disomy 3 to treatment by I-125 brachytherapy. The authors included 40 ciliochoroidal melanomas without matching them for size and location and observed that monosomy 3 tumors were statistically larger, especially ones located in the ciliary body and choroid at baseline. Furthermore, the larger tumors had a significantly greater decrease in thickness compared with the smaller tumors. This conclusion assumes that greater decrease in thickness is an indicator of more aggressive pathology, meanwhile the authors were comparing a broad spectrum of uveal melanomas that presents biases, such as size at baseline and location. In our study, the precaution to match tumor thickness among patients from the two subgroups was taken to avoid the bias of comparing shrinkage of larger and smaller tumors that could be proportionally different. Additionally, none of the other variables mentioned in the Table were statistically significant, even ones that were not intentionally matched in the two groups, such as patient age. The lack of statistical difference seen in all other variables makes the two subgroups matching acceptable for this comparative analysis. Interestingly, association with GEP class of the tumor cells does not appear on the basis of this study to explain the previously reported, unfavorable prognostic impact of faster tumor flattening following focal irradiation of posterior uveal melanomas. Our study is therefore consistent with the findings reported recently by Chappell and coworkers 6 following proton beam irradiation. 
The total number of cases and the length of recorded follow-up of our patients limit our ability to assess the independent prognostic significance of rate of post–plaque tumor flattening in multivariate analysis that includes GEP class of the tumor as a prognostic variable. Conversely, because this study showed no strong correlation between rate of post–plaque tumor flattening and the GEP class of the tumors, it is possible to speculate that previous published findings 14 have been a chance occurrence because of bias from tumor size, cell type, and other characteristics, which are intercorrelated and associated with GEP class. However, it is still possible, but less likely, that the rate of post-irradiation flattening may retain its prognostic significance even when GEP class is known. Larger and longer-term studies are needed to determine this possibility. 
Acknowledgments
Supported by an unrestricted grant from Research to Prevent Blindness, Inc., to the Department of Ophthalmology, University of Cincinnati, Cincinnati, Ohio (JJA), the Quest for Vision Fund and Ophthalmology Research and Education Fund of the University of Cincinnati, the Dr. E. Vernon and Eloise C. Smith Chair of Ophthalmology endowment fund (JJA), and Dr. Mary Knight Asbury Chair of Ophthalmic Pathology endowment fund (ZMC) of the University of Cincinnati. 
Disclosure: Z.M. Corrêa, None; J.J. Augsburger, None 
References
Augsburger JJ Gamel JW Shields JA Markoe AM Brady LW. Post-irradiation regression of choroidal melanomas as a risk factor for death from metastatic disease. Ophthalmology . 1987; 94: 1173–1177. [CrossRef] [PubMed]
Glynn RJ Seddon JM Gragoudas ES Evaluation of tumor regression and other prognostic factors for early and late metastasis after proton irradiation of uveal melanoma. Ophthalmology . 1989; 96: 1566–1573. [CrossRef] [PubMed]
Guthoff R Haase J von Domarus D Das Regressionsverhalten des Aderhautmelanoms nach Strahlentherapie – ein neuer prognosticher Parameter? Klin Mbl Augenheilk . 1990; 196: 6–10. [CrossRef] [PubMed]
Kaiserman I Anteby I Chowers I Post-brachytherapy initial tumour regression rate correlates with metastatic spread in posterior uveal melanoma. Br J Ophthalmol . 2004; 88: 892–895. [CrossRef] [PubMed]
Onken MD Worley LA Tuscan MD Harbour JW. An accurate, clinically feasible multi-gene expression assay for predicting metastasis in uveal melanoma. J Mol Diagn . 2010; 12: 461–468. [CrossRef] [PubMed]
Harbour JW Chen R. The DecisionDx-UM gene expression profile test provides risk stratification and individualized patient care in uveal melanoma. PLoS . 2013; 5.
Chappell MC Char DH Cole TB Uveal melanoma: molecular pattern, clinical features, and radiation response. Am J Ophthalmol . 2012; 154: 227–232. [CrossRef] [PubMed]
Onken MD Worley LA Char DH Collaborative Ocular Oncology Group Report Number 1: prospective validation of a multi-gene prognostic assay in uveal melanoma. Ophthalmology . 2012; 119: 1596–1603. [CrossRef] [PubMed]
Gist. Support vector machine. Overview of the SVM algorithm . Available at: svm.edsc.edu/svm-overview.html. Accessed September 19, 1999.
Augsburger JJ Gamel JW Bailey RS Donoso LA Gonder JR Shields JA. Accuracy of clinical estimates of tumor dimensions. A clinical-pathologic correlation study of posterior uveal melanomas. Retina . 1985; 5: 26–29. [CrossRef] [PubMed]
Marathe OS Wu J Lee SP Ocular response of choroidal melanoma with monosomy 3 versus disomy 3 after iodine-125 brachytherapy. Int J Radiat Oncol Biol Phys . 2011; 81: 1046–1048. [CrossRef] [PubMed]
Figure 1
 
Graph showing mean tumor thickness ± SD of the mean for tumor thickness at baseline, 3- and 6-months post irradiation for the posterior uveal melanomas in the two patient groups.
Figure 1
 
Graph showing mean tumor thickness ± SD of the mean for tumor thickness at baseline, 3- and 6-months post irradiation for the posterior uveal melanomas in the two patient groups.
Figure 2
 
Scatterplot graph showing tumor regression pattern by measurement of tumor thickness at baseline, 3-, and 6-months post irradiation for the posterior uveal melanomas in all 50 tumors studied.
Figure 2
 
Scatterplot graph showing tumor regression pattern by measurement of tumor thickness at baseline, 3-, and 6-months post irradiation for the posterior uveal melanomas in all 50 tumors studied.
Table
 
Comparison of Baseline Variables in GEP Class 1 and GEP Class 2 Patient Groups That Were Matched for Baseline LBD and Maximal Thickness of the Primary Posterior Uveal Melanoma
Table
 
Comparison of Baseline Variables in GEP Class 1 and GEP Class 2 Patient Groups That Were Matched for Baseline LBD and Maximal Thickness of the Primary Posterior Uveal Melanoma
Variable GEP Class 1 Categories GEP Class 2 Subgroup Chi-Subgroup df 2 P
Number (%) Number (%)
Age of patient, y
 ≤50 7 (28.0) 1 (4.0) 6.46 2 0.040
 >50 but ≤70 14 (56.0) 15 (60.0)
 >70 4 (16.0) 9 (36.0)
Sex of patient
 Male 12 (48.0) 9 (36.0) 0.74 1 0.39
 Female 13 (52.0) 16 (64.0)
LBD of tumor, mm
  ≤10 7 (28.0) 7 (28.0) 0.17 2 0.92
  >10 but ≤15 14 (56.0) 15 (60.0)
  >15 4 (16.0) 3 (12.0)
TH of tumor, mm
  ≤3.5 4 (16.0) 4 (16.0) 0.00 2 1.0
  >3.5 but ≤7 16 (64.0) 16 (64.0)
  >7 5 (20.0) 5 (20.0)
Intraocular tumor location
 Exclusively choroidal 19 (76.0) 19 (76.0) 0.00 1 1.0
 Involving ciliary body 6 (24.0) 6 (24.0)
Melanoma cell type
 Spindle 11 (44.0) 6 (24.0) 2.23 2 0.33
 Mixed 8 (32.0) 11 (44.0)
 Epithelioid 6 (24.0) 8 (32.0)
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×