Key words
breast - prostate - ultrasound
Introduction
Suspicion of a tumor
Suspicion of the presence of prostate cancer (PCa) results from a raised PSA level
and/or digital-rectal examination (DRE) [1]. Suspicion of the presence of mammary carcinoma results from a positive palpation
and/or suspicious imaging [2]. To either confirm or rule out the suspicion of cancer requires a histological examination
of punch biopsies of the relevant organs. In this case there is a significant difference.
Punch biopsy
Although numerous original studies emphasize that significant prostate cancer can
be reliably visualized using various imaging techniques, according to EAU guidelines,
randomized biopsy of the prostate using transrectal ultrasound (TRUS) is still currently
the gold standard [1]
[3]
[4]. As a rule, the ultrasound B-mode image is only used to detect the zonal anatomy
of the prostate, in order to subsequently obtain 10 – 12 “blind” tissue samples from
the predefined anatomical prostate zones. As the term “randomized” implies, TRUS is
not used to visualize the prostate cancer ([Fig. 1]). In contrast, when there is suspicion of breast cancer, a lesion visible in imaging
is biopsied selectively based on the image and not randomly biopsied ([Fig. 2]).
Fig. 1 Randomized prostate biopsy with removal of a core from the center right anatomical
zone without abnormalities in the B-mode image.
Fig. 2 Targeted breast biopsy with removal of a core from a suspicious region in the B-mode
image (arrow).
Palpation
Due to the increased cell and vessel density of carcinoma of the breast and prostate,
both types of tumors are harder than the normal surrounding tissue [5]. Although palpation of the breast can be performed by a woman herself is recommended
as precautionary measure [2], prostate self-examination by men is not performed, due to the less accessible location
of the prostate as well as other inhibitions. Furthermore, since only the portions
of the prostate accessible by finger lie against the rectal wall, anterior tumors
can escape detection in a clinical examination ([Fig. 3]). Due to both its exposed location as well as its deformability, the female breast
is not subject to the same limitations of palpation.
Fig. 3 Anterior tumor which escaped digital-rectal examination, yet in strain elastography
displays a region with increased tissue stiffness (coded blue, arrow).
Ultrasound strain elastography
Originally developed by Ophir et al. and introduced in 1991, USE was rapidly accepted
as a means to exhibit elasticity distribution in organs through imaging (color coded)
[6]. The advantage of this modern technique is that inaccessible and thus impalpable
hard – and therefore suspicious – changes can be detected ([Fig. 3]).
Technology
Technology
USE generates images with an image reconstruction rate greater than 30/sec during
which the organ is compressed and released using an ultrasound probe. Pressure on
the organ as well as the expected image quality can be read on an indicator on the
ultrasound image screen. After compression, hard tissue recovers at a different speed
compared to soft tissue, resulting in runtime differences of echo pairs, which, using
a mathematical algorithm, can be measured and then displayed color-coded as an elastogram
[7]. The color selection to represent the distribution of tissue elasticity can be arbitrarily
defined; for this article, hard regions of the breast are shown in grayscale (black = hard,
soft = white), for the prostate, a color scale is used (blue = hard, soft = red) ([Fig. 4]).
Fig. 4 Prostate cancer displayed as hard in the elastogram (left, coded blue, long arrow).
The short arrows indicate the hardened inner gland and thus not assessable in the
elastogram, likewise coded blue; breast cancer shown as hard in the elastogram (right,
coded black, long arrows).
Transducers
A linear probe is used to examine the breast to determine mammary carcinoma. To determine
prostate cancer, the prostate is examined using a transrectal probe; most working
groups use an end-fire probe, since a biplanar probe has limitations with respect
to this biopsy [8].
Macrocalcification
If macrocalcifications are present dorsally in the prostate, it is not possible to
examine the anterior prostate sections due to sound cancellation and limited alternative
access possibilities of the transrectal probe [3] ([Fig. 5]). The female breast offers a different situation, since macrocalcification-related
sound cancellation can be avoided by changing the position of the organ or the linear
probe.
Fig. 5 Macrocalcification of the prostate with noise cancellation and non-assessable anterior
sections.
Organ size
Examination of anterior sections of larger prostate organs is limited by the lack
of ductility due to anatomical conditions as well as the limited penetration depth
of the USE [9]. For the prostate there is an additional problem: the inner gland generally has
more connective tissue, is generally hypertrophic in the elderly and thus harder to
asses due to the greater basic hardness [10]. Although the female breast is made up of an individually highly variable combination
of glands, connective and fatty tissue, it appears more homogeneously in imaging,
and the USE limitation of penetration depth can be compensated for using organ compression,
in contrast to the prostate.
Tumor volume
Due to the tilt angle of the end-fire probe, the tumor diameter is correctly determined
in all 3 spatial directions only in apical prostate sections, whereas it is underestimated
based on height diameter [3] ([Fig. 6]). The linear transducer head does not have this problem, since it can flexibly probe
the target lesion. It has also been reported that USE can more accurately detect the
true size of the mammary carcinoma better than B-mode image ultrasound, since USE
also discloses the local invasion, which is not always visible in the B-mode scan
[11] ([Fig. 7]).
Fig. 6 In the large area section, prostate cancer shows a larger diameter (left, black arrows)
in the corresponding elastogram (right, white arrows).
Fig. 7 Two breast cancers displayed smaller in the B-mode image (left, arrow) than in the
corresponding elastogram (right, arrows).
B-mode image and elastogram
B-mode image and elastogram
Due to the loss of glandular architecture, an unambiguous image of prostate and breast
cancer shows a loss of reflexivity and therefore the B-mode image indicates a hypoechoic
tumor area [12]
[13]
[14]. Although some breast cancers can be iso- or hyperechogenic, the B-mode image indicates
a higher predictive value and high sensitivity to detection of breast cancer. This
contrasts with prostate cancer, since non-hypoechoic carcinomas are frequently encountered
[15]
[16]. This ultimately results in the B-mode image being used as a search modality for
breast cancer, and USE used only for additional evaluation/risk assessment [17]. For prostate cancer, USE is superior to the B-mode image with respect to detection,
and the B-mode image itself is only included in elastography scoring [18]
[19]. The B-mode image itself plays a role in measuring prostate volume and, as mentioned
earlier, is used in the identification of anatomical zones for the randomized biopsy
[12].
Strain ratio
The strain ratio (SR) was introduced since USE is an examiner-dependent freehand technique,
and additionally because benign changes such as inflammation or fibrosis with hardening
can affect the USE [20]
[21]. This is a semi-quantitative measurement procedure with which relative elasticity
values of lesions are generated. To do this, absolute strain values of elastographically
suspicious areas are compared in relation to elastographically unremarkable areas.
The ratio value thus calculated shows the degree of hardening compared to the unremarkable
tissue. Determining cut-off values was an attempt to obtain values that could distinguish
between benign and malignant lesions. The fat-to-lesion ratio (FLR) was established
in which the strain of the fat tissue (= reference tissue) of the breast is compared
the strain of the lesion. European studies determined cut-off values of 2.27 – 2.455,
depending on the technical equipment used, to differentiate between benign and malignant
focal lesions [20]
[21]. FLR appears to be a highly-significant additive tool for properly classifying a
lesion as BIRADS 3 or 4. For the prostate there is to-date only one published study
that investigated the value of strain ratio of prostate lesions [22]. Zhang et al. reported that the strain ratio could provide supplementary information
regarding prostate lesions in order to better differentiate between benign and malignant
lesions. In their study, a cut-off value of 17.44 was determined when the strain of
the lesion is compared to the strain of the contralateral side of the prostate used
as a reference. Our own unpublished results suggest that there is no stable reference
tissue available for the prostate, since the contralateral side can also be infiltrated
by infection or tumor. Furthermore, establishing a cut-off of the absolute strain
of the lesion seems to be more productive since a reference tissue is not required
in this case.
Physiology-Histology
Age
With increasing age, the breast becomes noticeably softer due to a lipomatous involution
of the mammary gland parenchyma [23]. An additional consequence is a loss of breast reflex which is generally reflected
in general hypoechoicity in the B-mode image. Under these circumstances it becomes
increasingly difficult to find hypoechoic breast cancer in the B-mode image due to
the lack of contrast in the hypoechoic background. Unlike examinations of the prostate,
USE plays an increasingly significant role in detecting breast cancer with advancing
patient age ([Fig. 8]). In contrast to the breast, with increasing age the prostate undergoes hypertrophy,
primarily of the inner gland, resulting in a reduction of elasticity and subsequent
hardening; this leads to a limitation of the elastographic detectability of prostate
cancer of these organ sections ([Fig. 9]). Chronic inflammatory processes of the prostate developed later in life additionally
aggravate this problem [10]. Consequently the B-mode image plays an increasingly important role in the detection
of prostate cancer.
Fig. 8 Involuted breast with poor B-mode image contrast, but with unambiguous hardened lesion
in the related elastogram.
Fig. 9 Hypertrophied prostate of an elderly man with completely hardened inner gland.
False-positive changes
Benign changes of the prostate such as inflammation, fibrosis, atrophy or adenomyomatosis
can involve increased tissue rigidity and can therefore be difficult to differentiate
elastographically from prostate cancer. This is certainly a reason for the occasionally
low positive predictive value of only up to 39 %, as we reported in a study of men
with PSA serum values < 4 ng/ML [24]. For most mammary lesions, benign tumors are harder than normal glandular tissues,
but softer than malignant lesions [25]. False-positive changes can be observed in hyaline fibroadenomas and fatty tissue
necrosis, since these can involve tissue hardening [5]
[26]
[27]. Additionally, USE has difficulty differentiating between postoperative scarring
and tumor recurrence, since in both cases there is reduced tissue elasticity. In this
instance contrast-enhanced MRI is superior to USE [17]
[28].
False-negative changes
If the prostate cancer tumor is made up of glands with dilated lumina containing substantial
mucus or contains sparse architecture, then such tumors can avoid detection by both
DRU as well as USE, due to increased tissue elasticity [29]
[30]. This type of histological tumor composition in generally found in prostate cancer
with primary Gleason pattern 3 (G6 (3 + 3) or G7 (3 + 4)). Above G7 (4 + 3), USE sensitivity
for prostate cancer detection is very high, since this cancer type is very compact
as a rule (dense architecture) [30]. Since prostate cancer ≤ G7 (3 + 4) is considered insignificant, and significant
if greater than G7 (4 + 3), USE particularly appears to be a possible technology to
reduce the much-criticized over-diagnosis and over-treatment of this cancer [31]. Since USE represents an additive to B-mode sonography, false-negative changes play
a reduced role in the detection of breast cancer. Most false-negative changes are
found, for example, in phyllodes tumors with a soft center, mucoid breast carcinomas
and carcinomas with large central necrosis [5]
[17]
[26]
[27]. In a B-mode image, both entities are shown as suspicious.
Tumor size
USE appears to play a significant role in the management of breast lesions < 5 mm
visible in the B-mode image, but not visible in mammography. Verification of reduced
elasticity of these lesions can result in a biopsy instead of only monitoring [25]
[26] ([Fig. 10]). In contrast, there is a finding of a suspicious tumor of the prostate if a hardened
area > 5 mm can be reproduced on two different planes [32]. For example, sensitivity of 9.7 % could be shown for prostate cancer with maximum
diameters of 0 – 5mm; 27 % for prostate cancer with maximum diameters of 6 – 10mm;
70.6 % for prostate cancer with maximum diameters of 11 – 20mm; and 100 % for prostate
cancer with maximum diameters of > 20 mm [30].
Fig. 10 Breast cancer < 5 mm in B-mode image (left, arrow), typically smaller than in the
elastography (right, arrow).
Summary
In contrast to diagnosis of the prostate, the B-mode ultrasound image is the leading
search method in diagnosing the breast, and USE is a useful additive examination technology.
Due to age-related changes in both the breast and prostate, the use of these technologies
appears to shift with the increasing age of the patient, however. Although there are
published multicenter studies regarding the value of USE in the diagnosis of breast
cancer, it would be desirable that the same were available for prostate cancer in
order to achieve a higher evidence level for this technology.
