February 2010
Volume 51, Issue 2
Free
Anatomy and Pathology/Oncology  |   February 2010
Intravascular Presence of Tumor Cells as Prognostic Parameter in Uveal Melanoma: A 35-Year Survey
Author Affiliations & Notes
  • Long V. Ly
    From the Department of Ophthalmology, Leiden University Medical Center (LUMC), Leiden, The Netherlands.
  • Omar F. F. Odish
    From the Department of Ophthalmology, Leiden University Medical Center (LUMC), Leiden, The Netherlands.
  • Didi de Wolff-Rouendaal
    From the Department of Ophthalmology, Leiden University Medical Center (LUMC), Leiden, The Netherlands.
  • Guy S. O. A. Missotten
    From the Department of Ophthalmology, Leiden University Medical Center (LUMC), Leiden, The Netherlands.
  • Gregorius P. M. Luyten
    From the Department of Ophthalmology, Leiden University Medical Center (LUMC), Leiden, The Netherlands.
  • Martine J. Jager
    From the Department of Ophthalmology, Leiden University Medical Center (LUMC), Leiden, The Netherlands.
  • Corresponding author: Martine J. Jager, Department of Ophthalmology, Leiden University Medical Center, PO Box 9600, 2300 RC Leiden, The Netherlands; m.j.jager@lumc.nl
Investigative Ophthalmology & Visual Science February 2010, Vol.51, 658-665. doi:10.1167/iovs.09-3824
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Long V. Ly, Omar F. F. Odish, Didi de Wolff-Rouendaal, Guy S. O. A. Missotten, Gregorius P. M. Luyten, Martine J. Jager; Intravascular Presence of Tumor Cells as Prognostic Parameter in Uveal Melanoma: A 35-Year Survey. Invest. Ophthalmol. Vis. Sci. 2010;51(2):658-665. doi: 10.1167/iovs.09-3824.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose.: Invasion of tumor cells into blood vessels is essential for metastasis of uveal melanoma. The occurrence of ingrowth of tumor cells in blood vessels in uveal melanoma was analyzed, and this parameter was compared with the survival of the patients.

Methods.: Between 1972 and 2007, 643 eyes primarily enucleated for uveal melanoma were evaluated histopathologically. Survival data were obtained from charts and from the Integral Cancer Center patient registry.

Results.: No vascular ingrowth of tumor cells occurred in 59% of the eyes, whereas 18% had tumor cell ingrowth in vessels inside the tumor, 10% in vessels outside the tumor, and 8% in vessels inside as well as outside the tumor. The presence of any intravascular ingrowth of tumor cells correlated significantly with the diameter (P < 0.01) and prominence of the tumor (P < 0.01), as well as with non-spindle-cell type (P = 0.03) and intrascleral ingrowth (P < 0.01), and was associated with a worse survival. When extravascular matrix patterns were not included in the multivariate analysis, intravascular ingrowth came out as an independent prognostic factor, but this was not the case when extravascular matrix patterns were included in the multivariate model.

Conclusions.: Intravascular ingrowth of tumor cells in uveal melanoma occurs frequently in combination with well-known histopathologic factors such as large tumor size, epithelioid cell type, and intrascleral ingrowth.

Uveal melanoma is the most common primary intraocular malignancy in adults, with an overall incidence of approximately six cases per million in the Western world. 1,2 Many different treatment modalities exist for the intraocular tumor, but these therapies seem not to be effective in preventing metastases, which occur in approximately 50% of patients. The average survival after diagnosis of metastases is only 10 to 18 months. 3 Because of a lack of lymphatics within the uveal tract, uveal melanomas disseminate predominantly hematogeneously. 4 Due to this fact, vascular invasion of tumor cells is a prerequisite for the formation of metastases. Blood vessels have been an important area of investigation during the past decades, as vessel growth inhibition has been found to prevent tumor growth in many types of cancer. 57  
The uvea is one of the most vascularized tissues of the human body, and blood vessels from the choroid are thought to extend into the uveal melanoma. In addition, it is assumed that the tumor itself creates blood vessels, 8,9 which can be identified by fluorescein angiography. 10,11 Histologically, the density of blood vessels can be evaluated, and specific areas with an increased microvascular density can be identified. Foss et al., 12 Mäkitie et al., 13 and Toivonen et al. 14 showed that a high microvascular density (MVD) in primary uveal melanoma tissue is associated with a worse survival. 
In certain tumors, including uveal and cutaneous melanoma, specific fluid-conducting channels have been observed, which are known as extravascular matrix patterns, or as vascular mimicry. 15 The presence of specific patterns such as loops and networks is associated with metastatic disease. These fluid-conducting meshworks lack endothelium and do not stain with the vessel markers CD 31 and CD 34, indicating that they are not regular blood vessels. Erythrocytes may travel through these channels, and one can thus imagine that tumor cells could also migrate via these extravascular matrix patterns toward larger blood vessels. 16  
The Leiden University Medical Center (LUMC) has been the main referral center for uveal melanoma in the Netherlands for almost 40 years. Since 1972, all enucleated eyes have been evaluated histologically by one pathologist according to a standard protocol. This analysis includes tumor cell infiltration into blood vessels inside the tumor, in transscleral vessels, and the vortex veins; however, the prognostic value of this specific parameter has hardly been evaluated. In the medical literature, very little information is available on this parameter. One study by our group mentioned this parameter, but this study was limited in number of specimens and time. 17 Coupland et al. 18 recently studied extraocular extension in relation to vortex veins. 
As mentioned, many studies show the importance of blood vessels in tumor growth 6,7,13,15 and in the development of metastases. This study was conducted to elucidate the importance of tumor cell ingrowth into blood vessels in uveal melanoma and to determine the prognostic importance of this phenomenon in a long-term study. 
Materials and Methods
Patients
Between August 1972 and August 2007, 715 eyes with uveal melanoma were enucleated in the Netherlands. The patients were observed, and when death occurred, the date was recorded. In addition to the patients' charts, the database of the Integral Cancer Center West was used, which registers data on metastases and checks the survival status of each patient with uveal melanoma on a yearly basis. In the Netherlands, cause of death is reported according to a standard protocol to the Central Bureau for Statistics (CBS, The Hague, The Netherlands). In addition, a specialized nurse registers information on clinical metastases or treatment for metastases. Follow-up time is measured in months. The data were updated in August 2007. 
The research protocol followed the current revision of the tenets of the Declaration of Helsinki (World Medical Association Declaration of Helsinki 1964; ethics for medical research involving human subjects). 
Pathology Specimens
Enucleation specimens were fixed in 4% buffered neutralized formalin. Between August 1972 and March 1995, enucleated eyes were embedded in celloidin after fixation for 48 hours and cut into 12-μm-thick mounted sections. The remaining sections were stored in 70% alcohol. From March 1995 onward, enucleated eyes were embedded in paraffin and serial sections of 4 to 5 μm were made of every globe and mounted on glass slides. From the embedded part of the eye, serial sections were made and every 10th section was mounted on a slide and examined. Some slides were stored for later diagnostic staining. Routine staining was performed with hematoxylin and eosin, and since January 1991, one periodic acid-Schiff (PAS) stain has been made of a representative section of each eye for examining extravascular matrix patterns. 
Histopathologic Examination
During 35 years, hematoxylin and eosin 12-μm celloidin- and 4-μm paraffin-embedded sections have been reviewed by one ocular pathologist (DdWR) in a standard fashion for confirmation of the diagnosis and evaluated for histologic parameters, including tumor location, largest basal diameter (in millimeters), prominence (apical height; in millimeters), cell type, intravascular ingrowth of tumor cells, intrascleral ingrowth, and the PAS-stained slides for examining extravascular matrix patterns, such as loops and networks. Largest basal diameter was determined by measuring the curve-shaped base of the tumor on the pathologic slide. Extravascular matrix patterns have been scored since January 1991. 15  
Intravascular tumor cell growth was classified into four categories: no ingrowth, ingrowth in vessels inside the tumor, ingrowth in vessels outside the tumor (i.e., in a intrascleral thin-walled blood vessel, sometimes reaching the ocular surface, like the vorticose vein; Fig. 1), and ingrowth in vessels both inside and outside of the tumor. Intrascleral tumor growth curling around emissary nerves and transscleral blood vessels, thereby reaching the ocular surface, was not considered intravascular tumor growth as these tumor cells did not invade the blood stream. 
Figure 1.
 
Intravascular ingrowth of tumor cells. (A) Uveal melanoma originating in choroid, ciliary body, and iris root. Tumor tissue is seen to penetrate through the episclera via the trabecular meshwork (T) and via the vorticose vein (V). Image of the original slide. (B) Uveal melanoma originating in the choroid. Tumor tissue in a sclera-perforating vorticose vein (outside the tumor). E, a tumor embolus. (C) Choroidal melanoma, epithelioid cell type. Melanoma cells (circle) and melanophage (arrow, M) within the lumen of a small blood vessel within the tumor. E, endothelial cells. Approximate magnification: (B) ×30; (C) ×120.
Figure 1.
 
Intravascular ingrowth of tumor cells. (A) Uveal melanoma originating in choroid, ciliary body, and iris root. Tumor tissue is seen to penetrate through the episclera via the trabecular meshwork (T) and via the vorticose vein (V). Image of the original slide. (B) Uveal melanoma originating in the choroid. Tumor tissue in a sclera-perforating vorticose vein (outside the tumor). E, a tumor embolus. (C) Choroidal melanoma, epithelioid cell type. Melanoma cells (circle) and melanophage (arrow, M) within the lumen of a small blood vessel within the tumor. E, endothelial cells. Approximate magnification: (B) ×30; (C) ×120.
The trabecular meshwork with its connection to Schlemm's canal and aqueous veins was considered vascular tissue; therefore tumor growth into the trabecular meshwork was scored as ingrowth of tumor cells in vessels outside the melanoma. 
Intrascleral ingrowth of tumor cells was classified into four different categories: no invasion of tumor cells, superficial (< one half of the sclera), deep (one half to three fourths of the sclera), and total scleral invasion. Episcleral and extraocular growth were included in the category total scleral invasion. 
COMS criteria were applied to reclassify the diameter and prominence of the tumors, as recorded in the original pathology report, into the three groups small, medium, and large. 19,20 The primary tumors were assessed on their pT categories (T1–T4) according the AJCC/UICC TNM classification (sixth edition). 21,22  
Data Analysis
ANOVA testing was used for comparing multiple groups (SPSS for Windows, ver. 12.0.1; SPSS, Chicago, IL). Tumor characteristics among categorized groups were compared with the χ2 test. 
Survival was assessed with the Kaplan-Meier survival analysis accompanied by the log rank test. Patients were censored when they died of a cause other than metastasis of uveal melanoma. Univariate Cox proportional hazards modeling was used to evaluate the prognostic value of the different histopathologic parameters. Besides visual inspection of the log-minus-log curves, we also performed Cox regression analyses including interaction terms of relevant covariates with time in the model to assess the proportionality of the hazards. In case when the proportional hazards seemed to be violated, we kept the interaction term in the multivariate model. Multivariate analysis identified the independent significant prognostic variables for survival. P < 0.05 was considered to be statistically significant. 
Results
Between 1972 and 2007, 715 eyes with a uveal melanoma were enucleated: 643 (90%) eyes were primarily enucleated, whereas 72 (10%) eyes had received prior treatment: TTT (transpupillary thermotherapy) in 11 (2%), ruthenium plaque therapy in 20 (3%), sandwich therapy (ruthenium plaque irradiation combined with TTT) in 28 (4%), and proton beam in 13 (2%) cases. 
We analyzed only the primarily enucleated eyes, since the group of eyes that had received prior treatment was relatively small and tumor characteristics may have been influenced by prior treatments. 
Of the 643 patients with enucleated eyes, 333 (52%) were men and 310 (48%) were women. Mean age was 58.8 years (SD ± 15.0 years). 
Tumors were classified as small in 14%, medium in 64%, and large in 21%, according to the COMS (Collaborative Ocular Melanoma Study) criteria and were mainly categorized as T2 (57%) or T3 (23%) according the TNM classification (6th edition). 21,22 The mean prominence was 5.7 mm (SD ± 3.2 mm) and the mean diameter 11.4 mm (SD ± 3.6 mm). 
Most tumors (59%) showed no ingrowth of tumor cells in any blood vessels, 18% of the patients had tumor cell ingrowth in vessels inside the tumor, 10% in vessels outside the tumor, and 8% in vessels both inside and outside the tumor. With regard to scleral ingrowth of tumor cells, superficial ingrowth (52%) occurred most frequently. 
The mean follow-up of patients was 8.8 years (SD ± 8.9 months), with a range of 0 to 31.6 years. Of all patients, 48% were alive at the last follow-up, 32% had died of metastases, and 21% had died of other causes. Eleven (2%) patients of the 643 were lost to follow-up. Other baseline characteristics can be found in Table 1
Table 1.
 
Baseline Characteristics of Patients and Histologic Data of Primarily Enucleated Eyes
Table 1.
 
Baseline Characteristics of Patients and Histologic Data of Primarily Enucleated Eyes
Numerical Variables Baseline Data Intravascular Ingrowth of Tumor Cells P
n % of n = 643* Total (n = 612) None† (n = 376) Inside† (n = 118) Outside† (n = 67) Both† (n = 51)
Sex
    Male 333 52% 319 61% 19% 11% 9% 0.96
    Female 310 48% 293 62% 20% 11% 8%
Eye
    Right 311 48% 298 61% 19% 12% 8% 0.66
    Left 322 52% 314 62% 20% 10% 9%
pTNM classification (6th edition)
    T1 84 13% 82 88% 5% 4% 4% 0.01
    T2 364 57% 348 53% 26% 13% 8%
    T3 150 23% 141 67% 16% 9% 8%
    T4 39 6% 36 56% 6% 11% 28%
Cilary body involvement
    Not present 514 80% 497 61% 20% 11% 8% 0.83
    Present 122 19% 114 63% 17% 11% 10%
Cell type
    Spindle 263 41% 252 67% 19% 8% 6% 0.03
    Mixed epithelioid 366 57% 281 57% 19% 13% 10%
Intrascleral ingrowth
    None 52 8% 51 90% 8% 0% 2% 0.01
    Superficial 331 52% 318 66% 21% 9% 4%
    Deep 127 20% 124 55% 21% 12% 12%
    Total sclera/episcleral 117 18% 110 44% 17% 20% 19%
Loops and/or networks
    Not present 76 12% 74 81% 5% 5% 5% 0.87
    Present 140 22% 139 77% 4% 9% 7%
Numerical Variables Baseline Data Intravascular Ingrowth of Tumor Cells P
None(n = 376) Inside(n = 118) Outside(n = 67) Both(n = 51)
Mean age, y 58.8 (±15.0) 57.2 (±15.0) 58.6 (±15.5) 63.0 (±13.8) 63.5 (±13.4) <0.01
Mean diameter, mm 11.4 (±3.6) 10.7 (±3.6) 11.7 (±2.9) 12.7 (±3.2) 14.1 (±4.4) <0.01
Mean prominence, mm 5.7 (±3.2) 5.3 (±3.3) 6.3 (±3.0) 6.1 (±2.7) 7.0 (±3.2) <0.01
Association of Intravascular Ingrowth of Tumor Cells with Other Histopathologic Parameters
Tumor cell ingrowth in vessels inside and outside the tumor was associated with a higher pT category and with scleral ingrowth (both P < 0.01; χ2 test): A deeper scleral penetration was associated with more frequent tumor cell invasion in any blood vessels (Table 1, Fig. 2). Furthermore, ingrowth of tumor cells in vessels was positively correlated with the presence of epithelioid cells (P = 0.03; χ2 test), higher age, a larger tumor diameter, and a greater tumor prominence (P < 0.001, P < 0.01, and P < 0.01 respectively, ANOVA test; Table 1). 
Figure 2.
 
Distribution of intravascular ingrowth versus intrascleral ingrowth.
Figure 2.
 
Distribution of intravascular ingrowth versus intrascleral ingrowth.
Survival Analysis
Kaplan-Meier analysis showed that patients with tumor cells in blood vessels had a significantly worse survival than did patients, who had no tumor cells in vessels (log rank test, χ2 = 31.5 and P < 0.01, Fig. 3 for the survival curve). 
Figure 3.
 
Kaplan-Meier survival curve of melanoma patients according to the different types of intravascular ingrowth. The numbers in the table represent the number of patients at risk.
Figure 3.
 
Kaplan-Meier survival curve of melanoma patients according to the different types of intravascular ingrowth. The numbers in the table represent the number of patients at risk.
Univariate Cox analysis demonstrated that intravascular ingrowth of tumor cells had a hazard ratio (HR) of 2.01 (P < 0.01) for vessels inside the tumor, of 2.29 (P < 0.01) with ingrowth in vessels outside the tumor, and of 2.49 (P < 0.01), if ingrowth occurred in both locations, compared with no vascular invasion of tumor cells. HRs for the other parameters can be found in Table 2
Table 2.
 
Cox Proportional Hazards Survival Analysis of Different Parameters with Death Due to Metastasis as the Endpoint
Table 2.
 
Cox Proportional Hazards Survival Analysis of Different Parameters with Death Due to Metastasis as the Endpoint
Cox Univariate
B-value P HR 95% CI
Sex, male/female* 0.24 0.09 1.27 0.96–1.67
Eye, right/left* −0.08 0.57 0.92 0.70–1.22
pTNM classification
    T1* 1
    T2 1.31 <0.01 3.69 1.99–6.84
    T3 1.79 <0.01 5.97 3.07–11.57
    T4 2.27 <0.01 9.67 4.50–20.76
Ciliary Body involvement, present/not present* 0.84 <0.01 2.31 1.69–3.17
Cell type
    Spindle* 1
    Mixed+epithelioid 1.05 <0.01 2.86 2.09–3.92
Intravascular ingrowth
    None* 1
    In vessels inside tumor 0.70 <0.01 2.01 1.44–2.81
    In vessels outside tumor 0.83 <0.01 2.29 1.52–3.46
    In vessels both inside/outside 0.91 <0.01 2.49 1.57–3.95
Intrascleral ingrowth
    None* 1
    Superficial 0.26 0.41 1.30 0.70–2.43
    Deep 0.60 0.07 1.83 0.95–3.53
    Total sclera 0.64 0.06 1.90 0.98–3.70
Loops and/or networks, present/not present* 1.65 <0.01 5.23 2.03–13.49
Age, for each year 0.03 <0.01 1.03 1.02–1.04
Largest basal diameter, for each mm 0.18 <0.01 1.20 1.16–1.24
Prominence, for each mm 0.13 <0.01 1.14 1.10–1.19
We also plotted survival graphs, in which patients were categorized according to specific prognostic factors and analyzed whether ingrowth of tumor cells in blood vessels led to a different survival within those categories (Fig. 4). Except with regard to the absence of loops and/or networks, ingrowth of tumor cells into blood vessels constituted a significant additional risk factor for mortality. 
Figure 4.
 
Kaplan-Meier survival curves for intravascular ingrowth after categorization according to known independent prognostic factors. P-value was obtained by the log rank test.
Figure 4.
 
Kaplan-Meier survival curves for intravascular ingrowth after categorization according to known independent prognostic factors. P-value was obtained by the log rank test.
Multivariate Analysis
Multivariate Cox regression analysis was performed to determine whether intravascular ingrowth is an independent significant prognostic factor for survival. Since extravascular matrix patterns have been scored since 1991, data are known concerning this parameter in 216 patients. We analyzed two multivariate models, one with and one without the presence of loops and networks as parameter. In the model without extravascular matrix patterns, ciliary body ingrowth (HR = 4.77, P < 0.01), largest basal diameter (HR = 1.02, P = 0.01), the presence of epithelioid cells (HR = 1.05, P = 0.04), age (HR = 1.03, P < 0.01), and intravascular ingrowth (HR = 1.60, P < 0.01) were independent prognostic factors for survival (Table 3A). If extravascular matrix patterns were added to this model, intravascular ingrowth turned out not to be a significant factor anymore, but ciliary body ingrowth (HR = 1.49, P = 0.02), largest basal diameter (HR = 1.11, P < 0.01), the presence of epithelioid cells (HR = 4.12, P < 0.01), and having extravascular matrix patterns “loops” and/or “networks” (HR = 4.43, P < 0.01) were independent predictive parameters for the survival of patients (Table 3B). 
Table 3.
 
Multivariate Analyses
Table 3.
 
Multivariate Analyses
A. Known Independent Prognostic Factors without Extravascular Matrix Patterns
B-value P HR 95% CI
Presence of ciliary body involvement 1.56 0.01 4.77 1.85–12.28
Presence of epithelioid cell type 0.49 0.04 1.05 1.01–1.10
Presence of intravascular ingrowth 0.47 0.01 1.60 1.17–2.17
Presence of intrascleral ingrowth NS
Age, for each year 0.03 0.01 1.03 1.01–1.04
Largest basal diameter, for each mm 0.02 0.01 1.02 1.01–1.03
Prominence, for each mm NS
B. Known Independent Prognostic Factors with the Extravascular Matrix Patterns Loops and/or Networks
B-value P HR 95% CI
Presence of ciliary body involvement 0.40 0.02 1.49 1.07–2.09
Presence of epithelioid cell type 1.42 <0.01 4.12 1.20–12.70
Presence of intravascular ingrowth NS
Presence of intrascleral ingrowth NS
Presence of loops and/or networks 1.49 0.01 4.43 1.46–13.48
Age, for each year NS
Largest basal diameter, for each mm 0.11 <0.01 1.11 1.03–1.20
Prominence, for each mm NS
Discussion
Several reports mention ingrowth of tumor cells into blood vessels in different types of malignancy, showing that the presence of tumor cells inside a blood vessel is associated with a worse survival. 2325 However, there is only a limited number of studies in uveal melanoma concerning ingrowth of tumor cells in blood vessels. 17,18 In our center, 715 eyes were enucleated for uveal melanoma over a 35-year time interval. Only the primarily enucleated eyes (n = 643) were analyzed. We did not include eyes with prior treatment, such as ruthenium-106 brachytherapy, in our study, due to the fact that the tumor characteristics could have been modified by the prior therapy. 26,27  
We observed that the presence of scleral invasion was associated with an increased frequency of tumor cell ingrowth in vessels inside and outside of the tumor. In general, melanomas use transscleral structures such as vessels and nerves to reach extraocular structures (e.g., the episclera): in those cases the tumor grows around the blood vessels and/or nerves. Regarding intravascular ingrowth: once tumor cells find their way inside a blood vessel, the bloodstream will carry the tumor cells to other parts of the body. Figure 1 demonstrates this situation: it can be observed that ingrowth of tumor cells into the vorticose vein occurred, which may then allow tumor cells to migrate outside the eye. 28 This can then be observed macroscopically as an extra large and dark vorticose vein on the sclera after enucleation. 
The presence of epithelioid cells is also associated with intravascular growth. In previous studies by Folberg et al., 29 and Vaupel and Gabbert 30 it was described that different cell types determined the growth pattern of normal vessels and extravascular matrix patterns in the tumor matrix. The different growth pattern of epithelioid cells compared to spindle cells can compress or even influence the constitution of the vascular lining, making it easier for tumor cells to invade the blood vessel. 
Folberg et al. 29 demonstrated previously with confocal microscopy that several patients, who had vascular loops in their tumors, lacked intravascular ingrowth of tumor cells and vice versa. Our larger study confirms this observation. 
Although 94 patients had no apparent ingrowth of tumor cells into blood vessels, they still developed metastasis. There are several possible explanations for this observation: one could be that we did not analyze the section in which tumor cells invaded a blood vessel. Another explanation could be that the tumor cells metastasized via the extravascular matrix patterns, and ingrowth into blood vessels is not absolutely needed. Another physiological explanation for this observed phenomenon could be that the tumor cells at one time invaded the blood stream, but after this process, the defect in the vessel wall was repaired. It may be that the histopathologic situation after enucleation is not representative of what previously happened in the tumor, since tumor cells may have been shed already, and the normal vessel structure restored. Schuster et al. 31 described that they could detect melanocyte-derived antigens with RT-PCR in peripheral blood samples of patients with uveal melanoma, and multivariate analysis showed this to be one of the most reliable independent prognostic markers for metastasis development. However, RT-PCR positive results were observed in only 10% of all cases, whereas in 25% of the patients' metastases occurred during the follow-up period. This result shows that an observation taken at one moment in time may not provide a complete picture. 
We also observed that ciliary body involvement was associated with scleral invasion. This phenomenon can be explained by the fact that tumors located in the ciliary body have a simple route for invading the sclera via the trabecular meshwork and so can easily infiltrate the episcleral structures. This finding is also in line with the literature: Seddon et al. 32 explained that contraction of the ciliary muscle facilitates the spread of tumor cells into the blood stream, leading to metastasis, and thus a bad prognosis. 
An additional factor that was related to ingrowth of tumor cells into blood vessels was the patient's age. In several articles, advanced age has been associated with a larger basal diameter, non–spindle-cell type and extrascleral extension, all of which are known prognostic factors for melanoma-related death. 3336 This was also the case in our study (data not shown). As tumor size and non–spindle-cell type were both clearly related to the presence of intravascular tumor cell ingrowth, which may explain the observed correlation with old age. 
Another explanation may be related to the innate immune system: recent findings in experimental models show a relation between age and blood vessel growth. 3739 Apte et al. 37 and Espinosa et al. 39 demonstrated that blood vessel growth in old mice differed from young mice: When retinal neovascularization was induced by laser, young mice showed limited neoangiogenesis, while old mice developed massive angiogenesis. Macrophages were thought to play a major role: proangiogenic M2-type macrophages are considered essential in helping tumor cells invade structures, such as blood vessels and the scleral matrix, by changing the integrity and construction of the tissue. Therefore, age may have important consequences for blood vessel growth and invasion of tumor cells into these structures, leading to more potential for metastatic spreading. 14 Of interest, the presence of a high number of macrophages in uveal melanoma has been found to be associated with tumor size, the presence of epithelioid cells, and other unfavorable prognostic factors, demonstrating the relevance of macrophages. 40  
Our data demonstrate that the presence of tumor cell invasion in blood vessels is associated with worse survival. Our survival results must be interpreted with care. Kujala et al. 41 described that survival could be overestimated with Kaplan-Meier analysis for patients with a long follow-up, since death related to melanoma after the first event contributes more to the estimate. However, multivariate Cox regression analysis demonstrated that if all well-known prognostic factors, such as ciliary body involvement, largest basal diameter, epithelioid cell type, and the presence of loops and/or networks are put into a model, all were independent prognostic factors, similar to results of Seregard and Kock. 42 Intravascular ingrowth is only an independent prognostic factor when extravascular matrix patterns are not added to the set of parameters analyzed. 
We show in this study that it is not necessary to add intravascular ingrowth of tumor cells to the set of prognostic parameters that need to be analyzed, since some better prognostic markers, such as extravascular matrix patterns are known. Although histopathologic parameters are important predictors of survival in uveal melanoma, chromosomal analysis 43,44 or microarray-based gene expression profiling 45 may be even more precise predictors. However, that histopathologic analysis is still useful was shown by Damato et al. 46 The best predictive index was not obtained from one test, but by using analysis of chromosome 3, basal tumor diameter, and cell type together. 
Footnotes
 Supported by the Netherlands Organization for Scientific Research NWO Mozaiek Grant 017.003.059 and KWF Grant 2001-2472.
Footnotes
 Disclosure: L.V. Ly, None; O.F.F. Odish, None; D. de Wolff-Rouendaal, None; G.S.O.A. Missotten, None; G.P.M. Luyten, None; M.J. Jager, None
References
Singh AD Topham A . Incidence of uveal melanoma in the United States: 1973–1997. Ophthalmology. 2003; 110: 956–961. [CrossRef] [PubMed]
Virgili G . Incidence of uveal melanoma in Europe. Ophthalmology. 2008; 114(12): 2309–2315. [CrossRef]
Kivela T Eskelin S Kujala E . Metastatic uveal melanoma. Int Ophthalmol Clin. 2006; 46: 133–149. [CrossRef] [PubMed]
Singh AD Borden EC . Metastatic uveal melanoma. Ophthalmol Clin North Am. 2005; 18: 143–150. [CrossRef] [PubMed]
Ferrara N Kerbel RS . Angiogenesis as a therapeutic target. Nature. 2005; 438: 967–974. [CrossRef] [PubMed]
Folkman J . How is blood-vessel growth regulated in normal and neoplastic tissue. Proc Assoc Cancer Res. 1985; 26: 384–385.
Folkman J . Tumor angiogenesis. Adv Cancer Res. 1985; 43: 175–203. [PubMed]
Boyd SR Tan DS de Souza L . Uveal melanomas express vascular endothelial growth factor and basic fibroblast growth factor and support endothelial cell growth. Br J Ophthalmol. 2002; 86: 440–447. [CrossRef] [PubMed]
Demou ZN Hendrix MJ . Microgenomics profile the endogenous angiogenic phenotype in subpopulations of aggressive melanoma. J Cell Biochem. 2008; 105: 562–573. [CrossRef] [PubMed]
Shields CL Shields JA De Potter P . Patterns of indocyanine green videoangiography of choroidal tumors. Br J Ophthalmol. 1995; 79: 237–245. [CrossRef] [PubMed]
Singh AD Rundle PA Rennie I . Retinal vascular tumors. Ophthalmol Clin North Am. 2005; 18: 167–176. [CrossRef] [PubMed]
Foss AJE Alexander RA Jefferies LW . Microvessel count predicts survival in uveal melanoma. Cancer Res. 1996; 56: 2900–2903. [PubMed]
Makitie T Summanen P Tarkkanen A Kivela T . Microvascular density in predicting survival of patients with choroidal and ciliary body melanoma. Invest Ophthalmol Vis Sci. 1999; 40: 2471–2480. [PubMed]
Toivonen P Makitie T Kujala E Kivela T . Microcirculation and tumor-infiltrating macrophages choroidal and ciliary body melanoma and corresponding metastases. Invest Ophthalmol Vis Sci. 2004; 45: 1–6. [CrossRef] [PubMed]
Folberg R Rummelt V Parys-Van Ginderdeuren R . The prognostic value of tumor blood-vessel morphology in primary uveal melanoma. Ophthalmology. 1993; 100: 1389–1398. [CrossRef] [PubMed]
Clarijs R van Dijk M Ruiter DJ de Waal RMW . Functional and morphologic analysis of the fluid-conducting meshwork in xenografted cutaneous and primary uveal melanoma. Invest Ophthalmol Vis Sci. 2005; 46: 3013–3020. [CrossRef] [PubMed]
Overkleeft ENM Zuidervaart W Hurks HMH . Prognostic value of the disodium phosphate P-32 uptake test in uveal melanoma: a long-term study. Arch Ophthalmol. 2003; 121: 1398–1403. [CrossRef] [PubMed]
Coupland SE Campbell I Damato B . Routes of extraocular extension of uveal melanoma: risk factors and influence on survival probability. Ophthalmology. 2008; 115: 1778–1785. [CrossRef] [PubMed]
The Collaborative Ocular Melanoma Study Group. Mortality in patients with small choroidal melanoma. COMS report no. 4. Arch Ophthalmol. 1997; 115: 886–893. [CrossRef] [PubMed]
Histopathologic characteristics of uveal melanomas in eyes enucleated from the Collaborative Ocular Melanoma Study. COMS report no. 6. Am J Ophthalmol. 1998; 125: 745–766. [CrossRef] [PubMed]
Greene FL Page DL Fleming ID . AJCC Cancer Staging Manual. 6th ed. New York: Springer; 2002: 193–200.
Sobin LH Witekind C . UICC TNM Classification of Malignant Tumors. 6th ed. New York: Wiley-Liss; 2002: 218–222.
Callejo SA Antecka E Blanco PL Edelstein C Burnier MNJr . Identification of circulating malignant cells and its correlation with prognostic factors and treatment in uveal melanoma: a prospective longitudinal study. Eye. 2007; 21: 752–759. [CrossRef] [PubMed]
Gabor S Renner H Popper H . Invasion of blood vessels as significant prognostic factor in radically resected T1–3N0M0 non-small-cell lung cancer. Eur J Cardiothorac Surg. 2004; 25: 439–442. [CrossRef] [PubMed]
Macchiarini P Fontanini G Hardin MJ . Blood-vessel invasion by tumor-cells predicts recurrence in completely resected T1–N0-M0 non-small-cell lung-cancer. J Thorac Cardiovasc Surg. 1993; 106: 80–89. [PubMed]
Kenneally CZ Farber MG Smith ME Devineni R . In vitro melanoma cell growth after preenucleation radiation therapy. Arch Ophthalmol. 1988; 106: 223–224. [CrossRef] [PubMed]
Kivela T Toivonen P . Analysis of melanoma cell type in uveal melanoma following treatment failure. Am J Ophthalmol. 2005; 140: 768–769. [CrossRef] [PubMed]
Raoof N Rennie IG Salvi SM . What is the significance of vortex vein invasion in uveal melanoma? Eye. 2009 8; 23(8): 1661–7. [CrossRef] [PubMed]
Folberg R Pe'er J Gruman LM . The morphologic characteristics of tumor blood vessels as a marker of tumor progression in primary human uveal melanoma: a matched case-control study. Hum Pathol. 1992; 23: 1298–1305. [CrossRef] [PubMed]
Vaupel P Gabbert H . Evidence for and against a tumor type-specific vascularity. Strahlenther Onkol. 1986; 162: 633–638. [PubMed]
Schuster R Bechrakis NE Stroux A . Circulating tumor cells as prognostic factor for distant metastases and survival in patients with primary uveal melanoma. Clin Cancer Res. 2007; 13: 1171–1178. [CrossRef] [PubMed]
Seddon JM Albert DM Lavin PT Robinson N . A prognostic factor study of disease-free interval and survival following enucleation for uveal melanoma. Arch Ophthalmol. 1983; 101: 1894–1899. [CrossRef] [PubMed]
Affeldt JC Minckler DS Azen SP Yeh L . Prognosis in uveal melanoma with extrascleral extension. Arch Ophthalmol. 1980; 98: 1975–1979. [CrossRef] [PubMed]
Chisholm JFJr . A long term follow-up on malignant melanomas of the choroid based on the Terry and Johns series. Am J Ophthalmol. 1953; 36: 61–75. [CrossRef] [PubMed]
Kidd MN Lyness RW Patterson CC Johnston PB Archer DB . Prognostic factors in malignant melanoma of the choroid: a retrospective survey of cases occurring in Northern Ireland between 1965 and 1980. Trans Ophthalmol Soc U K. 1986; 105: 114–121. [PubMed]
Shammas HF Blodi FC . Orbital extension of choroidal and ciliary body melanomas. Arch Ophthalmol. 1977; 95: 2002–2005. [CrossRef] [PubMed]
Apte RS Richter J Herndon J Ferguson TA . Macrophages inhibit neovascularization in a murine model of age-related macular degeneration. PloS Med. 2006; 3: 1371–1381. [CrossRef]
Dace DS Apte RS . Effect of senescence on macrophage polarization and angiogenesis. Rejuvenation Res. 2008; 11: 177–185. [CrossRef] [PubMed]
Espinosa-Heidmann DG Suner IJ Hernandez EP . Macrophage depletion diminishes lesion size and severity in experimental choroidal neovascularization. Invest Ophthalmol Vis Sci. 2003; 44: 3586–3592. [CrossRef] [PubMed]
Makitie T Summanen P Tarkkanen A Kivela T . Tumor-infiltrating macrophages (CD68(+) cells) and prognosis in malignant uveal melanoma. Invest Ophthalmol Vis Sci. 2001; 42: 1414–1421. [PubMed]
Kujala E Makitie T Kivela T . Very long-term prognosis of patients with malignant uveal melanoma. Invest Ophthalmol Vis Sci. 2003; 44: 4651–4659. [CrossRef] [PubMed]
Seregard S Kock E . Prognostic indicators following enucleation for posterior uveal melanoma: a multivariate analysis of long-term survival with minimized loss to follow-up. Acta Ophthalmol Scand. 1995; 73: 340–344. [CrossRef] [PubMed]
Scholes AG Damato BE Nunn J . Monosomy 3 in uveal melanoma: correlation with clinical and histologic predictors of survival. Invest Ophthalmol Vis Sci. 2003; 44: 1008–1011. [CrossRef] [PubMed]
Sisley K Rennie IG Parsons MA . Abnormalities of chromosomes 3 and 8 in posterior uveal melanoma correlate with prognosis. Genes Chromosomes Cancer. 1997; 19: 22–28. [CrossRef] [PubMed]
Onken MD Worley LA Ehlers JP Harbour JW . Gene expression profiling in uveal melanoma reveals two molecular classes and predicts metastatic death. Cancer Res. 2004; 64: 7205–7209. [CrossRef] [PubMed]
Damato B Duke C Coupland SE . Cytogenetics of uveal melanoma: a 7-year clinical experience. Ophthalmology. 2007; 114: 1925–1931. [CrossRef] [PubMed]
Figure 1.
 
Intravascular ingrowth of tumor cells. (A) Uveal melanoma originating in choroid, ciliary body, and iris root. Tumor tissue is seen to penetrate through the episclera via the trabecular meshwork (T) and via the vorticose vein (V). Image of the original slide. (B) Uveal melanoma originating in the choroid. Tumor tissue in a sclera-perforating vorticose vein (outside the tumor). E, a tumor embolus. (C) Choroidal melanoma, epithelioid cell type. Melanoma cells (circle) and melanophage (arrow, M) within the lumen of a small blood vessel within the tumor. E, endothelial cells. Approximate magnification: (B) ×30; (C) ×120.
Figure 1.
 
Intravascular ingrowth of tumor cells. (A) Uveal melanoma originating in choroid, ciliary body, and iris root. Tumor tissue is seen to penetrate through the episclera via the trabecular meshwork (T) and via the vorticose vein (V). Image of the original slide. (B) Uveal melanoma originating in the choroid. Tumor tissue in a sclera-perforating vorticose vein (outside the tumor). E, a tumor embolus. (C) Choroidal melanoma, epithelioid cell type. Melanoma cells (circle) and melanophage (arrow, M) within the lumen of a small blood vessel within the tumor. E, endothelial cells. Approximate magnification: (B) ×30; (C) ×120.
Figure 2.
 
Distribution of intravascular ingrowth versus intrascleral ingrowth.
Figure 2.
 
Distribution of intravascular ingrowth versus intrascleral ingrowth.
Figure 3.
 
Kaplan-Meier survival curve of melanoma patients according to the different types of intravascular ingrowth. The numbers in the table represent the number of patients at risk.
Figure 3.
 
Kaplan-Meier survival curve of melanoma patients according to the different types of intravascular ingrowth. The numbers in the table represent the number of patients at risk.
Figure 4.
 
Kaplan-Meier survival curves for intravascular ingrowth after categorization according to known independent prognostic factors. P-value was obtained by the log rank test.
Figure 4.
 
Kaplan-Meier survival curves for intravascular ingrowth after categorization according to known independent prognostic factors. P-value was obtained by the log rank test.
Table 1.
 
Baseline Characteristics of Patients and Histologic Data of Primarily Enucleated Eyes
Table 1.
 
Baseline Characteristics of Patients and Histologic Data of Primarily Enucleated Eyes
Numerical Variables Baseline Data Intravascular Ingrowth of Tumor Cells P
n % of n = 643* Total (n = 612) None† (n = 376) Inside† (n = 118) Outside† (n = 67) Both† (n = 51)
Sex
    Male 333 52% 319 61% 19% 11% 9% 0.96
    Female 310 48% 293 62% 20% 11% 8%
Eye
    Right 311 48% 298 61% 19% 12% 8% 0.66
    Left 322 52% 314 62% 20% 10% 9%
pTNM classification (6th edition)
    T1 84 13% 82 88% 5% 4% 4% 0.01
    T2 364 57% 348 53% 26% 13% 8%
    T3 150 23% 141 67% 16% 9% 8%
    T4 39 6% 36 56% 6% 11% 28%
Cilary body involvement
    Not present 514 80% 497 61% 20% 11% 8% 0.83
    Present 122 19% 114 63% 17% 11% 10%
Cell type
    Spindle 263 41% 252 67% 19% 8% 6% 0.03
    Mixed epithelioid 366 57% 281 57% 19% 13% 10%
Intrascleral ingrowth
    None 52 8% 51 90% 8% 0% 2% 0.01
    Superficial 331 52% 318 66% 21% 9% 4%
    Deep 127 20% 124 55% 21% 12% 12%
    Total sclera/episcleral 117 18% 110 44% 17% 20% 19%
Loops and/or networks
    Not present 76 12% 74 81% 5% 5% 5% 0.87
    Present 140 22% 139 77% 4% 9% 7%
Numerical Variables Baseline Data Intravascular Ingrowth of Tumor Cells P
None(n = 376) Inside(n = 118) Outside(n = 67) Both(n = 51)
Mean age, y 58.8 (±15.0) 57.2 (±15.0) 58.6 (±15.5) 63.0 (±13.8) 63.5 (±13.4) <0.01
Mean diameter, mm 11.4 (±3.6) 10.7 (±3.6) 11.7 (±2.9) 12.7 (±3.2) 14.1 (±4.4) <0.01
Mean prominence, mm 5.7 (±3.2) 5.3 (±3.3) 6.3 (±3.0) 6.1 (±2.7) 7.0 (±3.2) <0.01
Table 2.
 
Cox Proportional Hazards Survival Analysis of Different Parameters with Death Due to Metastasis as the Endpoint
Table 2.
 
Cox Proportional Hazards Survival Analysis of Different Parameters with Death Due to Metastasis as the Endpoint
Cox Univariate
B-value P HR 95% CI
Sex, male/female* 0.24 0.09 1.27 0.96–1.67
Eye, right/left* −0.08 0.57 0.92 0.70–1.22
pTNM classification
    T1* 1
    T2 1.31 <0.01 3.69 1.99–6.84
    T3 1.79 <0.01 5.97 3.07–11.57
    T4 2.27 <0.01 9.67 4.50–20.76
Ciliary Body involvement, present/not present* 0.84 <0.01 2.31 1.69–3.17
Cell type
    Spindle* 1
    Mixed+epithelioid 1.05 <0.01 2.86 2.09–3.92
Intravascular ingrowth
    None* 1
    In vessels inside tumor 0.70 <0.01 2.01 1.44–2.81
    In vessels outside tumor 0.83 <0.01 2.29 1.52–3.46
    In vessels both inside/outside 0.91 <0.01 2.49 1.57–3.95
Intrascleral ingrowth
    None* 1
    Superficial 0.26 0.41 1.30 0.70–2.43
    Deep 0.60 0.07 1.83 0.95–3.53
    Total sclera 0.64 0.06 1.90 0.98–3.70
Loops and/or networks, present/not present* 1.65 <0.01 5.23 2.03–13.49
Age, for each year 0.03 <0.01 1.03 1.02–1.04
Largest basal diameter, for each mm 0.18 <0.01 1.20 1.16–1.24
Prominence, for each mm 0.13 <0.01 1.14 1.10–1.19
Table 3.
 
Multivariate Analyses
Table 3.
 
Multivariate Analyses
A. Known Independent Prognostic Factors without Extravascular Matrix Patterns
B-value P HR 95% CI
Presence of ciliary body involvement 1.56 0.01 4.77 1.85–12.28
Presence of epithelioid cell type 0.49 0.04 1.05 1.01–1.10
Presence of intravascular ingrowth 0.47 0.01 1.60 1.17–2.17
Presence of intrascleral ingrowth NS
Age, for each year 0.03 0.01 1.03 1.01–1.04
Largest basal diameter, for each mm 0.02 0.01 1.02 1.01–1.03
Prominence, for each mm NS
B. Known Independent Prognostic Factors with the Extravascular Matrix Patterns Loops and/or Networks
B-value P HR 95% CI
Presence of ciliary body involvement 0.40 0.02 1.49 1.07–2.09
Presence of epithelioid cell type 1.42 <0.01 4.12 1.20–12.70
Presence of intravascular ingrowth NS
Presence of intrascleral ingrowth NS
Presence of loops and/or networks 1.49 0.01 4.43 1.46–13.48
Age, for each year NS
Largest basal diameter, for each mm 0.11 <0.01 1.11 1.03–1.20
Prominence, for each mm NS
×
×

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.

×