Hormone-secreting tumors are small and symptomatic
Although hormone-secreting tumors are not malignant in the strict sense of the definition, they should be mentioned here because of the high probability that they can be diagnosed by transvaginal sonography (TVS). These tumors are small, hiding at times in an ovary of almost normal size. They are also vascular, featuring a characteristic ring-like pattern, much like that of the corpus luteum, on color or power Doppler. They also produce general and clear clinical symptoms and signs. For example, testosterone-like tumors cause male-pattern baldness, hirsutism, and voice changes.
Many providers suspect a hormone-secreting tumor based on its signs and symptoms, and seek US confirmation from us. In many of these cases, laboratory tests have been done and point to the possible diagnosis—e.g., a high testosterone level in the case of a Sertoli-Leydig cell tumor.
One typical estrogen-secreting tumor is the granulosa cell tumor (FIGURE 3). This tumor can usually be identified by the solid-appearing tissue surrounding multiple cysts of different sizes; it is typically richly supplied with blood vessels.
Another clue to the diagnosis is the state of the endometrium. Because a granulosa cell tumor secretes estrogen, it causes a thickened endometrial lining and, usually, abnormal uterine bleeding.
FIGURE 3 Granulosa cell tumor
A. Sagittal image of the uterus demonstrating a thick, hyperechoic endometrial echo under hormonal stimulation of the tumor. B. Multicystic and solid areas alternate in the enlarged uterus. Power Doppler demonstrates the typical increased vascularity. (The arrows point to the cystic area of the tumor.)
Malignant ovarian neoplasms are rare
As a rule, the larger the lesion, the more suspicious it is.
Malignant tumors usually have a complex appearance:
- thick walls (≥4 mm)
- heterogeneous texture
- solid components
- papillary excrescences within the tumor as well as on the outer surface (FIGURE 4A and 4B).
FIGURE 4 Adenocarcinoma of the ovary
A. An enlarged right ovary containing several cystic structures. B. Right ovary and transverse section of the uterus. C, D. Power Doppler evaluation demonstrating rich vascularization. E. 3D orthogonal planes and volume calculation of the ovary (31.1 cc). F. 3D angiogram (lower right image) of the rich vascularization of the cancer. G. Relationship between the vascular right ovary and the uterus.Tumor vascularity is another marker suggestive of ovarian malignancy (FIGURE 4C and 4D). A fast-growing tumor requires a vascular “infrastructure,” a mesh of blood vessels that is laid down in expedited fashion and that is controlled by vascular growth factors. As explained in Part 2 of this series, the vessels in this vascular mesh lack the muscular layer of normal vessels. They frequently are intertwined, forming anastomoses and vascular lakes through which blood flows without much resistance. Look, therefore, for low resistance and high-velocity flow.
A new way to employ 3D US is to detect, measure, and quantify the blood supply to a tumor. FIGURE 4E shows how the vascularity and volume of an ovarian mass are calculated, with 3D angiographic display of the blood vessels contained within it demonstrated in FIGURE 4F. This vascular pattern can also be viewed in the context of the pelvic organs (FIGURE 4G), an approach that is useful in teaching.
Recently, Sladkevicus and colleagues used 3D US angiography to define tumor vascularity, identifying straight vessels, those that had changes in caliber, and bridging between vessels.1 They studied 104 patients who had 77 benign tumors, 6 borderline tumors, and 21 cancers. The researchers concluded that dense vessel patterns in the tumor made malignancy five times more likely. Widely dispersed straight vessels without branching were the strongest predictors of benign status, reducing the likelihood of malignancy by a factor of 10.1
We described the importance of a finding of blood vessels in an internal papillary structure as an accurate predictor of malignancy. We focused on a small volume of the mass, which was selected by a software program, and found that a preselected volume of 1 cc could reliably predict an increased, and pathological, vascular supply to an ovary containing cancer.2,3
Although ovarian cancer is rare, affecting 30 to 50 women of every 100,000, it is particularly deadly, with a 5-year survival rate (all stages) of 50%. If cancer is detected and treated during stage I, the 5-year survival rate rises substantially—to 95%. Sadly, only 25% of cases are detected while the cancer is still localized.
In stages III and IV, the 5-year survival rate is 28% or lower. It has been estimated that, if 75% of patients had their cancer detected during stage I, the mortality rate could be halved.
The lifetime risk of ovarian cancer in a woman who has no affected relative is 1.4% (1 case in every 70 women). When the patient has one affected first-degree relative, that risk rises to 5% (1 case in 20 women), and it rises to 7% (1 case in 14 women) when she has two or more affected first-degree relatives.