Noninvasive, in vivo, characterization of cutaneous metastases using a novel multimodal RCM-OCT imaging device: A case-series

Abstract

Background:

Cutaneous metastases (CM) are clinically challenging, requiring an invasive biopsy for confirmation. A novel, RCM-OCT device combines the advantage of high-resolution reflectance confocal microscopy (RCM) images and deeper optical coherence tomography (OCT) images to aid in non-invasive diagnosis.

Objective:

Characterize CM using RCM-OCT device.

Methods:

Seven patients with suspicious CM were consented and imaged with RCM-OCT device. CM features were defined by comparing with histopathology. Tumor depths were measured on OCT and on H&E-images and correlated using Pearson test. 3D-OCT images were reconstructed to enhance tumor visualization.

Results:

6/7 lesions were CM, and one was vascular ectasia, on histopathology. CM appeared as grayish-darkish oval to round structures within the dermis on RCM and OCT-images. On RCM, individual tumor cells were seen, enabling identification of even small tumor foci; while, on OCT deeper tumors were detected. Inflammatory cells, dilated vessels, and coarse collagen was identified in the dermis. Pearson correlation had an r² of 0.38 and a significant p-value <0.004 for depth measurements. CM could be differentiated from ecstatic vessels on 3D reconstructed OCT-image.

Limitation:

Small sample size and lack of clinical mimickers.

Conclusion:

RCM-OCT can detect CM and has potential in aiding noninvasive diagnosis and management.

Cutaneous metastases (CM) are rare and clinically challenging¹, often requiring a biopsy for confirmation. Biopsies are invasive and can lead to complications. Noninvasive imaging could aid in diagnosis and management of the CM. Techniques such as dermoscopy², ultrasound³, and reflectance confocal microscopy (RCM) have been explored to characterize CM but had limitations. Recently, a novel multi-modal non-invasive, in vivo imaging RCM-OCT device that combines the advantage of RCM and optical coherence tomography (OCT) by providing RCM cellular-level, high-resolution images in the horizontal (en face) plane (200 μm depth) and OCT low-resolution images in the transverse plane (1000 μm depth) has been developed⁴ and shown to help detect BCC⁵.

In this study, we used the RCM-OCT device to image and describe features of CM in breast cancer patients to elucidate its potential role in aiding their diagnosis and management. Seven patients with clinically suspicious cutaneous metastases from breast cancer were imaged. Multiple en face RCM stacks and cross-sectional OCT rasters were acquired for each lesion. Following RCM-OCT imaging, a biopsy was performed for H&E analysis. Images were analyzed by two experts to describe morphology and cytology of the tumor cells. Secondary features associated with CM at various layers of the skin were also noted. In addition, morphometric analysis of tumor depth on OCT images was performed by measuring the distance of 3–5 superficial tumor foci per lesion to the top of the epidermis, to correlate with H&E tumor cell depth. RCM and OCT descriptive and morphometric findings were correlated with H&E. 3D reconstruction of OCT images was performed for two cases to help differentiate common ectatic vessels from CM.

Histopathology confirmed CM in six of the seven lesions imaged. On RCM, tumor was identified in 4/6 lesions. In all these lesions, metastases were seen in the dermis at a varying depth. They appeared grayish to darkish well-demarcated, oval to round clusters of varying sizes. Within these tumor clusters, small to medium sized cells with a dark nucleus and variable amount of grayish cytoplasm were seen (Figs. 1–3). Sparse to mild inflammation, seen as small bright cells, was identified in all lesions. Similar to the findings on RCM, CM were recognized only in the dermis on OCT. However, metastases were identified in 5/6 lesions on OCT due to increased depth detection compared to RCM. Metastases appeared grayish to dark, hypo-reflective well-demarcated (most) to ill-defined (one) tumor nodules of varying sizes and shapes in the superficial dermis. Pearson correlation test showed a direct linear correlation between OCT depth and histopathology depth, with an r² of 0.38 and a significant p-value of <0.004 (Supplemental Material 4–5).

Figure 1: Features of cutaneous metastasis from breast cancer on clinical, RCM, and OCT images and their comparative histopathology.

Case 1: Clinical image shows an erythematous 20 mm plaque on left breast, clinically suspicious for CM (black arrow). On RCM, gray (hyporeflective) lobulated tumor nodules (yellow asterisk) composed of medium-sized cells with grayish cytoplasm and dark nucleus are seen within the dermis. Intra-tumoral vessels are seen (red arrow within the tumor nodule). Surrounding dermis, has coarse collagen (green star) and dilated vessels (red arrows). On OCT, these tumor nodules (encircled by yellow dotted circles) appear grayish in color (yellow asterisk) with surrounding dense white collagen (green star) and dark round-tubular dilated vessels (red arrows). H&E image: On histopathology, the lesion was diagnosed as an invasive, moderately to poorly ductal carcinoma of breast. The tumor (encircled by yellow dotted circles) has a micro papillary pattern and is composed of variable-sized pleomorphic tumor cells. The stroma has scar-like dense collagen (green star) and a few ectatic vessels (red arrows). Case 2: Clinical image shows an erythematous 30 mm nodule on posterior head, suspicious for CM (black arrow). RCM image shows grayish-dark (hyporeflective), medium, round-oval nodules (yellow asterisks) composed of small grayish cells, within the dermis. In the surrounding dermis has mildly coarse collagen (green star), small round prominent vessels (red arrows), and scattered inflammatory cells (blue arrow). On OCT, these nodules (encircled by yellow dotted circles) appear as oval, grayish-dark structures (yellow asterisk) surrounded by whitish and dense collagen (green star) and dark round-tubular prominent vessels (red arrow). H&E image: On histopathology, the lesion was diagnosed as metastatic matrix-producing neoplasm, morphologically similar to patient’s prior breast tumor. The tumor nodules (encircled by yellow dotted circles) show pushing borders and are composed of small tumor cells with scant cytoplasm seen in a myxoid stroma. Few dilated lymphatics and small proliferative vessels (red arrow) are seen in the mildly sclerotic dermal collagen (green star). RCM image FOV= 0.75mm², OCT FOV= 1 mm x 2mm, H&E image =20X.

Figure 3: Images from two cutaneous metastatic lesions from breast cancer showing advantages of a combined RCM-OCT device.

Case 4 with cutaneous metastasis only visible on RCM: a) RCM image shows small grayish-dark (hyporeflective) tumor foci (yellow asterisk) with grayish tumor cells. The surrounding dermis has coarse collagen (green star) and dilated vessels (red arrow). H&E image in the inset shows tumor (yellow arrows) as linear clusters or single cells with a monomorphic appearance, which were seen only on the RCM image. These tumor foci were barely visible on OCT image (not shown in the the figure). Case 5 with cutaneous metastasis only visible on OCT: Clinical image shows a brown 7 mm papule on the left abdomen, suspicious for CM (black arrow). a) OCT image show prominent multiple round-oval grayish (hyporeflective) tumor nodules (encircled by yellow dotted circles and yellow asterisks) in the deeper dermis. The dermis had coarse collagen (green star) and dilated linear vessels (red arrow) around these tumor nodules. On histopathology (inset), the lesion was diagnosed as a metastatic carcinoma and had tumor (encircled by yellow dotted circles) with a micropapillary pattern in the reticular dermis. Few prominent vessels (red arrow) and mild sclerosis (green star) are seen in the dermis. Due to deeper location of the tumor, cutaneous metastasis was not visible on RCM image. RCM image FOV= 0.75mm², OCT FOV= 1 mm x 2mm.

RCM-OCT could successfully identify CM in all the lesions by combining the high-resolution images of RCM and deeper depth images of OCT. Study limitations include a small sample size, sampling bias, and lack of control groups. We envision that the combined technique of RCM-OCT could be the most effective mode to diagnose cutaneous metastases, to guide biopsy if necessary and to monitor them non-invasively during treatment.

Figure 2. A: Case 3: Features of cutaneous metastasis from breast cancer on clinical, RCM, and OCT images and their comparative histopathology.

A) RCM image acquired from a skin-colored 10 mm papule on the right breast, clinically suspicious for CM (black arrow; inset) shows a well-demarcated tumor nodule (hyporeflective) composed of medium-sized grayish cells with dark nuclei (yellow asterisk), within the dermis (green star). Plump bright cells (orange arrows) are seen within the tumor nodules (yellow asterisk). The collagen (green star) appears thickened and organized around the tumor nodule (yellow asterisk). Few dilated linear vessels are visualized in the dermis (red arrow).

Figure 2B: On OCT, these well-demarcated nodules appear as round to oval hyporeflective structures (encircled by yellow dotted circles) with grayish-white material (yellow asterisk) within them. The dermis (green star) shows coarse collagen (green star) and dilated vessels (red arrow) around the tumor nodules. On histopathology, the lesion was diagnosed as an invasive carcinoma consistent with carcinoma of mammary origin. H&E image (inset) shows large tumor clusters (encircled by yellow dotted circles) composed of large pleomorphic cells immediately underneath the epidermis. The plump bright cells within the tumor nodule on RCM correspond to melanophages (orange arrows). The stroma has mild sclerosis (green star) and a few prominent vessels (red arrow). RCM image FOV= 0.75mm², OCT FOV= 1 mm x 2mm, H&E image =20X.

Data Availability Statement:

Raw data that supports the findings of this study are available from the corresponding author, upon reasonable request.