Survival after primary enucleation for choroidal melanoma: changes induced by the introduction of conservative therapies

Outcomes of fractionated CyberKnife radiosurgery in patients with choroidal malignant melanoma

OBJECTIVE

The aim of this study was to evaluate local tumor control and complication development rates of fractionated CyberKnife radiosurgery (CRS) in patients with choroidal melanoma.

METHODS

A total of 29 patients with choroidal melanoma were treated with fractionated CRS at Ankara Oncology Research and Training Hospital, Department of Radiotherapy between May 2009 and December 2013. Patients were treated with CRS if the initial height of the choroidal melanoma was ≥ 6 mm, or juxtapapillary and/or juxtamacular tumors with a height of >2.5 mm. Ophthalmic examinations were performed at baseline and at months 3, 6, 9 and 15 after radiotherapy. Assessment of visual acuity and measurement of tumor base dimension and height using A-scan and B-scan echography were done at each visit.

RESULTS

The mean age was 56 (27-75) years. Tumor was located on choroid in 23 and on ciliochoroid in 6 patients. 86.2% of all melanomas were classified as medium sized and 23.8% as large sized. A median total dose of 5000 cGy was applied. Median tumor height decreased from 7.5 mm at baseline to 4.4 mm at the last follow-up visit (p < 0.001). Median visual acuity decreased from 0.4 at baseline to hand motion (p < 0.001). One patient had been lost to the metastatic disease and one patient had been treated with enucleation due to recurrent tumor growth.

CONCLUSION

CRS is an effective and reliable local treatment modality in uveal melanoma.

[The prognostic value of uveal melanoma cell type]

OBJECTIVE

To determine the prognostic value of a uveal melanoma (UM) cell type for the development of metastases (MTS).

SUBJECTS AND METHODS

The investigation enrolled 96 patients (96 eyes) with UM after enucleation. Forty-one patients without signs of MTS were included in this group, who were followed up for more than 36 months (mean, 70.5 months (36 to 105 months)), 55 patients with MTS who were followed up for an average of 21 months (2 to 44 months). The MTS and non-MTS groups were statistically homogeneous in age, gender, tumor size, location, and ciliary body involvement in the process, as well as in extrabulbar growth.

RESULTS

There were spindle cell, mixed cell, and epithelioid cell UMs in 44, 35, and 21% of patents, respectively. The tumors in patients without MTS were noted to be significantly more likely to have spindle cell-type UM (p<0.0001). Mixed cell and epithelioid cell UMs were more frequently detected in patients with MTS (p<0.0001), which was believed to be due to the presence of epithelioid cells in both cell types. A survival analysis showed that the 3- and 5-year survival rates for patients with spindle cell UM were significantly higher than that for those with epithelioid cell or mixed cell UM (p<0.001); the 3-and 5-year survival rates for spindle cell UM were 78 and 70%, respectively; those for mixed cell UM were 37 and 24%; and those for epithelioid cell UM were 50 and 31%.

CONCLUSION

The similar survival rates for patients with mixed cell or epithelioid cell type UM could conclude that it is advisable to use the binary principle - the presence or absence of epithelioid cells in the tumor, when assessing a UM cell type as a prognostic factor.

[Management of uveal melanomas, guidelines for oncologists]

Uveal melanomas are the most frequent primary malignant eye tumor. Enucleation was historically the gold standard. Since then, several studies showed that conservative treatments did not increase the risk of metastasis or survival. Choroidal melanomas are both radioresistant and located close to visual structures (the optic nerve and macula) of the eye, which may be preserved in some settings without compromising tumor control, as this is the first priority. Different types of radiation therapy may be used for such tumors: brachytherapy and charged particles, including proton beam therapy. If visual prognosis is dependent to the local treatment, the vital prognosis is dependent on the metastatic risk, with a risk of liver involvement in 20 to 50% of patients, depending on tumor size and genomics. Median survival after the discovery of liver metastases is about 15 months. The management of these patients is often complex. Systemic therapies (chemotherapy, targeted therapies, immunotherapy, etc.) yield limited response rates and although local treatments of liver metastases are promising, they are only feasible in selected patients. The mission of the MELACHONAT national network is to improve the management of patients regardless of the stage of the disease. The patient association ANPACO is dedicated to help uveal melanoma patients in their health care path and to promote knowledge dissemination within the patient community. The aim of this review is to focus on the local treatments of uveal melanomas as well as the management of their metastatic evolution.

Feasibility of Image-Guided Robotic Radiotherapy using Three Fractions for Uveal Melanoma

Aims

A retrospective study was performed to demonstrate the feasibility and efficacy of 3-fraction image-guided robotic stereotactic radiotherapy (fSRT) for uveal melanoma.

Materials and Methods

Six patients with medium-sized or large tumors, who declined enucleation, were enrolled. The gross tumor volume (GTV) ranged from 454 to 2185 mm3. The total doses included 36 or 39 Gy in 3 fractions.

Results

Follow-up ranged from 19 to 40 months. In 5 patients, the tumor mass gradually underwent an average 24.5% size reduction. All 3 patients with a GTV <1000 mm3 had a functional eye, while 3 patients with a GTV ≥1000 mm3 did not have a functional eye. Radiation-induced complications occurred to some degree in all patients. However, complications that required enucleation were not detected.

Conclusion

We suggest that image-guided robotic radiotherapy using 3 fractions is a feasible and safe treatment option for patients with uveal melanoma. In cases of medium-sized and large tumors, fSRT could be used as an alternative treatment for cases ineligible for brachytherapy, but a longer follow-up and a larger number of patients are required to confirm the suitability of the method.

Risk factors associated with secondary enucleation after fractionated stereotactic radiotherapy in uveal melanoma

PURPOSE

To evaluate risk factors for secondary enucleation after fractionated stereotactic radiotherapy (fSRT) in uveal melanoma.

METHODS

In this retrospective study, clinical data of 118 consecutive patients who had initially been treated with fSRT between 1999 and 2009 were collected and analysed. The patients who had undergone secondary enucleation were identified and examined for clinical, histopathological and cytogenetical (fluorescence in situ hybridization determined) data. Also, the reasons for secondary enucleation, such as treatment failure (progressive tumour growth or tumour recurrence) or complications following fSRT (painful blind eye), were recorded and examined.

RESULTS

The secondary enucleation rate was 16% after a median follow-up of 4.7 years, with 5% due to treatment failure and 11% due to complications. In the univariate analysis, large tumour diameter (p = 0.019) and large tumour height (p = 0.001) were associated with secondary enucleation, tumour involvement of the optic disc showed borderline significance (p = 0.068). Cox regression multivariate analysis displayed large tumour height as independent prognostic factor (HR 1.42, 95% CI 1.12-1.81, p = 0.004). Following secondary enucleation, mitotic figures were present in five of 18 tumours, and gain of chromosome 8q was also present in five tumours. Within the subgroup of patients who required secondary enucleation due to failed tumour control by fSRT (N = 6), mitotic figures were present in four of six tumours while gain of 8q was present in three of six tumours.

CONCLUSION

Secondary enucleation after previous fSRT was associated with large tumour height. High mitotic counts and gain of chromosome 8q were frequently found in secondary enucleations and possibly indicate a more aggressive or radiation-resistant tumour.

The comparison of ruthenium brachytherapy and simultaneous transpupillary thermotherapy of choroidal melanoma with brachytherapy alone

PURPOSE

To compare the outcomes of combined treatment of choroidal melanoma with ruthenium brachytherapy (BT) simultaneously with transpupillary thermotherapy (TTT) and treatment with BT alone.

METHODS AND MATERIALS

Two matched groups of patients, one treated with BT and simultaneous TTT (Group BT+TTT, n=63), the other treated with BT alone (Group BT, n=70) were analyzed retrospectively. The main outcome measures were rate of tumor regression, recurrences, enucleations, metastases, recurrence-free and overall survival rate, and visual acuity, assessed by Kaplan-Meier analysis.

RESULTS

Patients were matched according to mean age (p=0.22), mean tumor thickness (6.4 vs. 6.25mm, range 2.5-10.8mm, p=0.59), and mean length of followup (42 vs. 34.4 months, range 3-109, p=0.052). Tumor largest basal diameter (13.0 vs. 12.9mm), tumor location, and mean radiation dose (apical 135 vs. 136Gy and scleral 1294 vs. 1438Gy) were also similar in both groups (p>0.1). Treatment with BT+TTT resulted in higher rate of tumor regression (63% vs. 49%, respectively, p=0.036), lower 5-year tumor recurrence rate (96% vs. 83%, p<0.034), and higher eye-globe preservation (98% vs. 87%, p<0.024) and recurrence-free survival rates (89% vs. 67%, p<0.017) than treatment with BT alone. There was no difference in complications (p>0.5), metastasis-free (93% vs. 81%, p>0.22) and overall survival rates (91% vs. 81%, p>0.39), or in visual outcomes.

CONCLUSION

Combined treatment of choroidal melanoma with ruthenium BT and simultaneous TTT seems to provide higher local control, eye-globe preservation, and recurrence-free survival rates than treatment with BT alone and results in similar rates of metastases and overall survival.

Initial results of fractionated CyberKnife radiosurgery for uveal melanoma

OBJECTIVE

To evaluate prospectively local tumor control and morbidity after fractionated CyberKnife radiosurgery for uveal melanoma unsuitable for ruthenium-106 brachytherapy or local resection.

METHODS

This study includes melanoma > or =7 mm in initial height, or juxtapapillary and/or juxtamacular tumors (height > or =3 mm; posterior tumor margin extending to within 3 mm of optic disk rim and/or fovea). Patients were excluded if they presented evidence of echographic extrascleral tumor extension, neovascular glaucoma, or any form of pretreatment or metastases at baseline. The eye was stabilized by the same ophthalmologist via peribulbar injection of 5 cc 2% lidocaine. CyberKnife radiosurgery was performed delivering a total dose of 60 Gy to the 80% or 85% isodose line in three fractions. The planning target volume (PTV) included the contrast-enhancing lesion on MRI plus a 1-mm margin (no margin on fovea site).

RESULTS

Five patients with uveal melanoma were treated by this procedure. All patients had serous retinal detachment associated with the tumor. No grade > or =2 acute toxicities were observed. Eight-month follow-up revealed a decrease in tumor thickness in three patients and reattachment of the retina in four. The tumors remained stable in two eyes and an increase in retinal detachment was noted in one eye. Vision improved minimally in two eyes and remained stable in three.

CONCLUSION

CyberKnife fractionated radiosurgery seems to be a viable alternative local treatment modality in uveal melanoma with no serious acute side effects. Further follow-up is indicated.

Expression and distribution of MMPs and TIMPs in human uveal melanoma

Matrix metalloproteinases (MMPs) and tissue inhibitors of MMPs (TIMPs) are involved in tumour invasion, metastasis and angiogenesis, and have been implicated as progression markers in uveal melanoma, although their topographical expression has not been fully described. In this study we compared the distribution and specificity of several classes of MMPs (MMP-1, -2, -9, -19, and MT1-MMP) and physiological MMP inhibitors (TIMP-2 and -3) in different regions of the tumour microenvironment and adjacent choroid in a series of primary uveal melanomas. Paraffin sections of untreated uveal melanomas (n=18, 3/18 spindle; 11/18 mixed, and 4/18 epithelioid) were examined for MMP-1 (collagenase 1), MMP-2 and MMP-9 (gelatinases A and B), MT1-MMP (membrane-type 1-MMP), MMP-19, TIMP-2 and TIMP-3 (tissue inhibitors of MMPs), using indirect peroxidase immunohistochemistry. The distribution and intensity of immunolabelling was graded semi-quantitatively (0-3) by 2 independent observers. Non-parametric analyses were used to test for associations between tumour cell type, and the average grade of MMP or TIMP expression. Immunostaining for MMP-1, -9, -19 and MT1-MMP was > or =Grade 2 in more than 70% of specimens, and a heterogeneous pattern of MMP-1, -9, MT1-MMP and TIMP-3 expression was observed. At the tumour-scleral interface (TSI), melanoma cells had a flattened morphology and a much reduced MMP and TIMP expression, with a high expression in tumour areas adjacent to the TSI. Tumour vasculature and stromal cells strongly expressed MMP-2. We also observed heterogeneous immunostaining of the vasculature by MMP-1, -9, MT1-MMP and TIMP-2 antibodies, and of the extravascular matrix by MMP-9 antibody. The distinct immunostaining patterns observed for MMPs and TIMPs within uveal melanoma are consistent with their involvement in tumour growth and angiogenesis. In particular, the heterogeneous expression within regions of the tumours, and the localized expression in vasculature and stromal cells emphasises the importance of the tumour microenvironment in the pathogenesis of uveal melanoma (and other tumours).

Expression and distribution of MUC18 in human uveal melanoma

The immunoglobulin superfamily protein MUC18 is involved in transendothelial migration and signal transduction, and is expressed in malignancies including cutaneous melanoma. Recent in vitro studies showed evidence of increased MUC18 protein in some uveal melanoma cell lines with an increased potential for invasion. We assessed seven uveal and three metastasis-derived melanoma cell lines for the expression of MUC18 mRNA and protein by RT-PCR, and immunoblotting and immunocytochemistry, respectively. We also examined the expression and distribution of MUC18 in paraffin sections of primary uveal melanomas (n = 23; 5/23 spindle; 18/23 mixed and epithelioid) and normal eyes (n = 3) using a polyclonal goat anti-human antibody to MUC18 visualized with peroxidase and Vector NovaRED. Distribution and intensity of immunostaining was graded semi-quantitatively (grade 0 to 3) by 2 independent observers. All cell lines expressed MUC18 mRNA and protein ( approximately 130 kDa), and showed punctate cell membrane MUC18 immunostaining. Primary melanomas displayed heterogeneous cell membrane and cytoplasmic MUC18, with moderate to strong immunolabelling (> or =grade 2) in approximately 70% of tumours. Vasculature in tumours and in retina and choroid of all melanoma-affected and normal eyes showed intense MUC18 immunostaining. These observations further suggest a role for MUC18 in uveal melanoma growth; moreover, interactions between MUC18-positive melanoma cells and vasculature may be important for the hematogenous spread of cells during metastases.