OA Imaging – Pathology Correlation
Opto-acoustic imaging is a new technology that may be able to demonstrate anatomic tumor morphology and provide functional information about tumor vascularity and deoxygenation without the need for contrast injection or ionizing radiation.
OA color maps demonstrate an increased total amount of hemoglobin as well as greater deoxygenation of hemoglobin within and/or surrounding malignant breast tumors.
The distribution of these OA features within the tumor interior versus the external boundary zone and periphery correlates with the histologic grade of the malignancies.
The histopathologic differences between high and low grade cancers reveal the basis for the OA findings.
Low-grade carcinomas are paucicellular with most of the internal tumor volume comprised of desmoplasia and/or extracellular matrix.
The smaller number of tumor cells present are only mildly dedifferentiated.
Grade 1 invasive duct carcinomas have relatively few biologically active immune cells.
Correspondingly,
OA reveals few internal vessels with little or no deoxygenation (Figure 7).
Externally,
low-grade malignancies demonstrate prominent desmoplasia and peripheral radiating vessels.
These histologic features correspond to architectural distortion mammographically and,
while difficult to appreciate with conventional ultrasound,
are dramatically displayed on OA color maps resulting in a high external OA score.
Conversely,
high-grade carcinomas are composed of larger numbers of relatively more dedifferentiated tumor cells and also have many biologically active immune cells that rapidly deoxygenate the vascular tumor interior.
OA images demonstrate profuse vascularity and deoxygenation within these tumors,
revealing their biologic aggressiveness (Figure 6).
Unlike lower grade malignancies,
they are characterized by less desmoplasia and few peritumoral radiating vessels resulting in a low external OA score.
These aggressive carcinomas more often present as circumscribed masses without associated architectural distortion,
overlapping in appearance with benign lesions on anatomic imaging,
but tend to have distinctive features on OA imaging due to their high internal vascularity and deoxygenation.
The strong inverse correlation between the total external OA score and biologic aggressiveness may also reflect host immune response.
The boundary zone surrounding the growing tumor is frequently a site of an intense and metabolically active immune response.
In particular,
tumor- associated lymphocytes and macrophages may be present in significant numbers with a consequent positive or negative influence on the biologic aggressiveness of the neoplasm [3].
This peritumoral immune response may contribute to altered depictions of external vascularity and deoxygenation.
Thus,
a higher external score in prognostically favorable,
lower grade malignancies suggests a favorable tumor immune response and merits further study,
which would require correlation with numbers and types of leukocytes within the external zones.
OA and Existing Functional Breast Imaging Modalities
As the radiology community strives to continue to improve the accuracy of breast imaging,
interest in functional imaging has grown.
Elastography has been proposed as a method of improving the accuracy of ultrasound diagnosis on the basis of measuring tumor stiffness.
Two forms currently exist,
known as strain and shear-wave elastography. In a multinational study,
Berg et al.
showed that shear-wave elastographic features can be used to selectively downgrade BI-RADS 4a masses and upgrade BI-RADS 3 masses improving specificity from 61.1% to 78.5% [4].
The wide adoption of elastography in clinical practice has been hindered by a number of factors.
Inter- and intra-observer variability are relatively high due to dependence on the degree of applied pressure.
Elastographic features are less reliable for lesions deeper than 2 cm.
Finally,
significant false positive and false negative rates are observed due to the presence of soft malignancies and hard benign lesions with reported sensitivities as low as 81.7% for strain elastography and 86% for shear-wave elastography [5-10].
Breast MRI has offered a valuable adjunctive modality based on tumor angiogenesis.
While highly sensitive,
MRI is limited in its specificity due to the preponderance of vascular benign lesions [11-14].
Furthermore,
it requires contrast injection,
carries a high cost,
and excludes patients with contraindications to exposure to a magnetic field.
Scintimammography,
another potentially helpful functional modality,
is similarly limited by false positive benign lesions,
the need for radionuclide injection,
and a relatively high total body radiation dose compared to conventional mammography [14,
15]. In addition to the logistic advantages of being relatively low cost without the need for contrast injection or ionizing radiation,
OA technology takes functional information to a level beyond imaging the vascularity of tumors,
which may be high in both malignant and benign lesions.
It further provides a measure of the oxygenation within the vessels of these tumors,
which may help to distinguish between benign vascular lesions that remain oxygenated and malignancies which rapidly extract oxygen.
Limitations
Our study has several limitations.
It is an early study based on a small number of patients.
The analysis of histopathologic features was based primarily on H&E stains.
Furthermore,
much of the neovascularity and deoxygenation in breast malignancies is within the boundary zone rather than the tumor interior.
Conventional pathologic sections do not optimally represent the external boundary and peripheral zones.
Follow-up studies using large section or megacassette histopathologic correlation to better define the causes of external findings,
are planned.
Future Directions
These findings are based on a feasibility study. Opto-acoustic imaging is currently undergoing further investigation to validate this data.
A prospective multi-institutional pivotal trial in the U.S.
(PIONEER) recently completed active enrollment of over 2,000 subjects with follow-up data to be collected by August 2015. Additionally,
a post-market clinical follow-up study is soon to be underway in the Netherlands.
The future of OA imaging may potentially include several diverse applications.
Aside from improving accuracy in the interpretation of solid breast masses,
the technology may play a role in assessing response to neoadjuvant chemotherapy,
evaluating axillary lymph nodes,
and directing management of high-risk lesions.
Furthermore,
it may prove useful in organs other than the breast,
such as the evaluation of tumors in the prostate or thyroid glands.
Further study is needed to elucidate the underlying mechanisms of OA imaging and its potential benefits as a functional imaging platform.