Ex vivo high-definition optical coherence tomography of

JEADV
DOI: 10.1111/jdv.12063
ORIGINAL ARTICLE
Ex vivo high-definition optical coherence tomography of
basal cell carcinoma compared to frozen-section
histology in micrographic surgery: a pilot study
T. Maier,* D. Kulichová, T. Ruzicka, C. Kunte,† C. Berking†
Department of Dermatology and Allergology, Ludwig-Maximilian University of Munich, Munich, Germany
*Correspondence: T. Maier. E-mail: [email protected]
Abstract
Background Micrographic surgery is an established, but time-consuming operating procedure for facial basal cell
carcinoma (BCC). A new high-definition (HD) optical coherence tomography (OCT) with high lateral and axial
resolution in a horizontal (en-face) and vertical (slice) imaging mode allows a fast and non-invasive in vivo
examination of BCC.
Objectives To compare the diagnosis of BCC in excised tissue ex vivo by high-definition optical coherence
tomography (HD-OCT) with the findings of frozen-section histology in micrographic surgery.
Methods Twenty freshly excised BCC were examined by HD-OCT in the en-face and slice imaging mode divided
into four sections each in concordance with the four excision margins of histography, and subsequently processed
for conventional micrographic evaluation.
Results A total of 80 HD-OCT images of 20 BCCs were evaluated and in 45% (9 ⁄ 20) HD-OCT correlated perfectly
with the histography results. The sensitivity and specificity for the 80 evaluated HD-OCT images were 74% and 64%
respectively.
Conclusions High-definition optical coherence tomography allows the postoperative identification of BCC in
excised tissue ex vivo, but has still limitations in the recognition of tumour margins in comparison with the
micrographic evaluation of frozen sections.
Received: 6 August 2012; Accepted: 9 November 2012
Conflict of Interest
The high-definition optical coherence tomography device used in this study was provided by AGFA Healthcare
GmbH. Dr. Maier has served as lecturer ⁄ consultant for AGFA Healthcare GmbH.
Funding Sources
This work was supported by the Curd-Bohnewand-Fonds of the University of Munich (to T.M.), by the Matthias
Lackas Foundation and the Dr. Helmut Legerlotz Foundation (to C.B.).
Introduction
Basal cell carcinoma (BCC) is the most prevalent tumour in the
USA, Australia and Europe1 with rising incidence.2,3 The mean
age for developing BCC is 60 years, although studies show a rising
incidence also in younger people of around 40 years. Tanning bed
use and smoking have been shown to be independent risk factors
for BCC.4
†
Authors share senior coauthorship.
This work is dedicated to Prof. Hans-Christian Korting who was
always a source of indispensable knowledge and has recently passed
away.
JEADV 2012
Although it is a usually slowly growing and extremely rarely
metastasizing tumour of the skin, the exact and complete excision
of BCC is critical because of the locally destructive behaviour and
the high propensity to recur.
For this reason, micrographic surgery has been developed to
reduce the recurrence rate of BCC, mainly in high-risk areas such
as the face.5,6
Different approaches of micrographic surgery have been
described such as the widely used Mohs surgery, the margin strip
method (‘Tübinger Torte’) and the ‘Munich method’. These
methods differ with regard to the technique of excision and preparation of the tumour tissue, but they have all in common the complete tumour removal with tumour-free margins at all sites to
ª 2012 The Authors
Journal of the European Academy of Dermatology and Venereology ª 2012 European Academy of Dermatology and Venereology
Maier et al.
2
prevent local recurrence and the optimal conservation of tumourfree surrounding tissue to achieve the optimal cosmetic
outcome.7–10 However, regardless of the method micrographic
surgery is time-consuming, laborious and expensive.
In this context, the application of new imaging techniques such
as reflectance confocal microscopy and optical coherence tomography (OCT) may potentially facilitate the process of evaluating
the tumour tissue immediately after excision. In a recent study,
OCT was applied ex vivo for the detection of BCC before the tissue
was processed for frozen sections for Mohs micrographic surgery
and demonstrated a low specificity (56%) and sensitivity (19%).11
Recently, a new high-definition (HD)-OCT device has been
developed, which in addition to the vertical OCT imaging mode
offers a real-time horizontal ‘en-face’ imaging mode, which allows
the immediate visualization of OCT pictures with high resolution
in both dimensions. It could be shown recently that high-definition optical coherence tomography (HD-OCT) is useful in the
detection of non-melanoma skin cancer such as BCC and actinic
keratosis.12,13 In the ‘Munich method’ of micrographic surgery,
the frozen tissue is also cut in horizontal sections and thus facilitating a correlation with the horizontal en-face imaging mode of
HD-OCT.
The purpose of this study was to evaluate the feasibility of the
innovative HD-OCT in the ex vivo analysis of BCC tumour margins in comparison with the histographic analysis of the tumours
as the gold standard.
(speckling). According to the producer’s instruction, the penetration depth in skin in optimal conditions can be about 450–
750 lm. The resolution is 3 lm in all three dimensions according
to the producer’s instruction. Skintell can work in two different
modes: real-time b-scan (slice) and c-scan (en-face) and additionally allows fast capture of a 3D tomogram. The OCT probe is
applied directly onto the skin with an optical gel (Skintell optical
gel, AgfaHealthCare, Belgium) as coupling medium. The field of
view in the en-face mode is 1.8 · 1.5 mm.
The clinically suspicious lesions were systematically evaluated by
HD-OCT in the slice and en-face mode and images were recorded.
The lesions were examined by an experienced investigator (T.M.)
prior to histographic evaluation and blinded to the histopathological result. The HD-OCT was scored positive for BCC whenever
typical morphological features of BCC islands (dark ⁄ grey lobulated nodules with a dark rim) were found in the evaluated
HD-OCT image as described previously by our group.12
Micrographic surgery and histopathology
After the HD-OCT imaging was performed, the tissue was routinely processed for micrographic surgery according to the
‘Munich method’ as described previously.8 This method includes
serial sections of the lesion in a horizontal way of sectional cutting.
The histological evaluation of the frozen-cut and haematoxylineosin-stained tissue was performed by a dermatologist, experienced in the evaluation of histographic tissue, blinded to the
HD-OCT result (C.K.).
Patients and methods
We evaluated 20 randomly selected biopsy-proven BCCs immediately after the first excision of micrographic surgery by HD-OCT.
The margins of the excised fresh tissue were divided into four
regions and labelled (I–IV) by the surgeons, and this was documented on a corresponding sketch. HD-OCT was then performed
ex vivo placing the tissue flat on a petridish with the epidermis
upside focusing on the four documented margins of the excised
tissue. Multiple en-face scans were obtained at steps of 3 lm in
depth. The margin was evaluated as positive if characteristic signs
of BCC as described previously were recognized within the imaging field. In addition, an HD-OCT image of the centre of the
excised lesion was taken. Subsequently, the tissue was processed as
usual for the micrographic evaluation.
High-definition optical coherence tomography (HD-OCT)
The commercially available full-field HD- OCT system (Skintell,
AgfaHealthCare, Belgium) was used for OCT images. The system
is based on a time-domain OCT system with dynamic focus tracking, which is a synchronized motion of the imaging lens system
and the reference optical system. An ultra-high-speed infrared
camera is used, which allows a high scanning speed especially, in
the 3D mode. The light source is a halogen lamp with a Gaussian
filter and an ultra-high bandwidth centred at 1300 nm (infrared
light) with high depth resolution and reduced lateral crosstalk
JEADV 2012
Statistical analysis
Descriptive statistics was performed to evaluate sensitivity and
specificity for ex vivo HD-OCT images in reference to the histological results and P-values were determined using Chi-squared test.
Statistical analysis was performed using the statistical package SPSS
(Superior Performance Software System, Munich, Germany). A
P-value of < 0.05 was regarded as significant.
Results
The 20 excised BCCs were of the following subtypes: 13 nodular,
four infiltrative, one ulcerated, one nodular-infiltrative and one
unclassified. The tumours were located on the nose (n = 8), on or
close to the ear (n = 5), periocular (n = 4), on the forehead or
temple (n = 2) and in the perioral area (n = 1) (Table 1).
Four HD-OCT images per tumour were analysed according to
the surgical margins (I, II, III, IV), thus 80 images were evaluated
altogether.
The ex vivo HD-OCT analysis of freshly excised BCCs showed
typical tumour formation with grey ⁄ dark nodules often surrounded by a darker rim and bright peritumoural stroma as previously described.12 The en-face (horizontal) HD-OCT (Table 2)
images of tumourous areas were matched with the horizontal sections of histography as exemplarily shown in Figs 1 and 2. The
quality of ex vivo imaging did not significantly differ from HD-
ª 2012 The Authors
Journal of the European Academy of Dermatology and Venereology ª 2012 European Academy of Dermatology and Venereology
Basal cell carcinoma in ex vivo high definition OCT
3
Table 1 Clinical and histopathological characteristics
Gender
Male
11
Female
9
Age range (years)
53–90
Tumour site
Nose
8
Ear
5
Periocular
4
Forehead ⁄ temple
2
Perioral
1
Histopathological type
Nodular
13
Infiltrative
4
Ulcerated
1
Nodular-infiltrative
1
Unclassified
1
OCT images of in vivo tissue except for tumour shrinkage and collapsed vessels; tumour islands were still recognizable as shown in
Figs 1 and 2, but with slightly reduced contrast compared to in
vivo tissue (data not shown). The HD-OCT imaging procedure
required about 2 min per slice and en-face images, which resulted
in a processing time of 8 min for four images per tumour.
In nine of 20 (45%) examined tumours, the HD-OCT examination correlated exactly with the histography in all four marginal
regions (Table 2). Four BCCs showed corresponding results
between HD-OCT and histography in three of the four investigated
margins, two BCCs in two of four margins, and three BCCs in only
one of four margins. In two tumours, none of the four tumour
margins was evaluated correctly by HD-OCT (Table 2). The results
gained by HD-OCT were not influenced by the BCC subtype.
Evaluating the 80 HD-OCT results of the marginal pictures, the
sensitivity of HD-OCT in the diagnosis of BCC was 74% and the
specificity was 64%. The investigated series showed a positive predictive value of 61%.
Discussion
Previously it has been shown that OCT is a possible tool for the
imaging of BCC. The possibility to measure the tumour thickness
of BCC by OCT was investigated in different studies.14–19 Recently,
we could show that HD-OCT with the new en-face mode and a
high lateral and axial resolution of 3 lm is a valuable tool in the
diagnostics of BCC.12
(a)
(b)
(c)
Figure 1 HD-OCT image in slice (a) and en-face mode (b, micrometre bar 200 lm) of BCC no. 17 showing the tumour margin no. II.
In the slice mode (a, arrow) a darker roundish nodule is depicted at the right lateral part of the image consistent with a BCC tumour
nodule as shown in the histography margin no. II (c, arrows). In the en-face mode (b, arrows), several darker oval structures are present close to the right lateral margin of the image in consistency with the histographic image (c, arrows) where blue tumour nodules
and strands are found (haematoxylin–eosin staining, x2, micrometre bar 100 lm).
JEADV 2012
ª 2012 The Authors
Journal of the European Academy of Dermatology and Venereology ª 2012 European Academy of Dermatology and Venereology
Maier et al.
4
(a)
(b)
(c)
Figure 2 In the high-definition optical coherence tomography (HD-OCT) slice (a) and en-face mode (b, micrometre bar 200 lm) of
BCC no. 19, greyish tumour strands are displayed destroying the regular architecture of the skin. The histographic image (c, arrows)
shows the same tumour formations as depicted in the HD-OCT image with antlers-like blue tumour formations (arrows, haematoxylin–eosin staining, x2, micrometre bar 100 lm). This HD-OCT image was evaluated as a margin still including tumour parts in congruence with the tumour-positive histography.
Table 2 Correlation of ex vivo HD-OCT and micrographic histology (MGH) in excised basal cell carcinoma (BCC, n = 20). The
table shows the number of BCCs and the respective number of
tumour margins (0–4) correctly identified by HD-OCT
Number of MGH margins
Number of correctly identified
BCC by HD-OCT
4
3
2
1
0
9
4
2
3
2
In this study, the feasibility of ex vivo HD-OCT analysis of BCC
was evaluated in comparison with the histographic procedure as
the gold standard.
As HD-OCT offers the possibility of the horizontal (en-face)
imaging mode, there was a good correlation with the horizontally
prepared sections for the histological evaluation. We recognized
similar features of the tumour such as buds and nodules in both
HD-OCT and histology.
High-definition optical coherence tomography is a fast method
in contrast to the time-consuming histography, but it lacks sufficient sensitivity and specificity. One main limitation of our study
is that only four images of each tissue sample (one per marginal
area) had been analysed and thus tumour-containing areas might
have been missed. In the HD-OCT device, the imaging field is
JEADV 2012
restricted to 1.8 · 0.5 mm and the penetration depth is limited to
450–750 lm in optimal conditions, whereas other OCT devices
possess a scan area of 5 · 5 · 2 mm and a detection depth of
about 1–2 mm. The limited penetration depth of HD-OCT does
not allow a reliable evaluation of the tumour thickness in vivo.
Although the advantage of HD-OCT compared to the conventional OCT is a higher resolution of about 3 lm and the possibility of a real-time en-face scanning mode, there are conventional
OCT devices available with the option of 3D images but not, to
our knowledge, in a real-time manner.
The advantage of the ex vivo imaging is the possibility to image
the tissue upside down and thus detecting the deeper layers of the
excised tumour. This technique has not been performed in this pilot
study, but might be an interesting goal for further studies. In consistence with Cunha et al.11 we noticed in the ex vivo scanning a
shrinkage of the tissue and collapse of the vessels, but tumour
islands were still visible as shown in Figs 1 and 2. To diminish the
loss of image quality, a preoperative tumour scanning in vivo could
be an option in further studies, although to date the size of the HDOCT probe does not allow an exact labelling of tumour margins.
Here, a mapping guided by the dermoscopic image as it is possible
in reflectance confocal microscopy would be helpful. Alternatively,
the screening of remnants of tumour nests in the operating field
after surgery could be a promising approach in the future.
ª 2012 The Authors
Journal of the European Academy of Dermatology and Venereology ª 2012 European Academy of Dermatology and Venereology
Basal cell carcinoma in ex vivo high definition OCT
In a recent study, OCT was applied ex vivo in the detection of
BCC on frozen sections prepared for Mohs micrographic surgery
and resulted in a low specificity (56%) and low sensitivity
(19%).11 The authors speculated that further technical development might enhance the predictiveness of ex vivo OCT in the evaluation of BCC. In this study, we could show for the first time that
HD-OCT with the possibility of the en-face scanning mode was
able to reach a higher sensitivity (74%) and specificity (64%). But
nevertheless, in the majority of cases, HD-OCT was not sufficient
for a precise diagnosis of all BCC tumour margins as seen in
micrographic surgery. Here, further technical developments would
be necessary to reach the precision of histography, but without the
time-consuming cutting and staining of the tissue.
Although there are multiple studies on OCT analysis of nonmelanoma skin cancer, there is only one other study on eight
periocular BCCs, which were examined ex vivo with an en-face
OCT. In that study, three OCT devices were compared, but none
of them reached the high resolution of HD-OCT. Nevertheless,
similar to our findings, the dark tumour nodules of BCC separated clearly from the surrounding tissue by highly reflective
structures and could thus easily be identified.20 Due to the low
sample number, no calculation of sensitivity or specificity had
been performed.
Another promising imaging technique is the ex vivo reflectance
confocal microscopy (RCM), which is currently studied for the
evaluation of tumour margins of different skin tumours. It has
already been shown that RCM is a valuable tool in the recognition
of BCC21–23 and some studies have demonstrated the ex vivo RCM
analysis of BCC.24–26 Schüle et al. examined 66 BCCs after preparation for histography by ex-vivo RCM, and found a sensitivity of
42% and a specificity of 77% in lateral tumour margins.27 Despite
of the high resolution of RCM of about 1.5 lm, it has not been
established routinely in the ex vivo diagnostics so far, although it is
already very commonly used in the in vivo diagnosis of melanoma
and non-melanoma skin cancer.28–30 For the in vivo examination,
there is a limited penetration depth of about 250 lm, which does
not allow a complete evaluation of the tumours in depth whereas
in ex vivo imaging, the scanning process is not limited to a certain
depth and is comparable with routine histology. One limitation is
the quite time-consuming scanning and evaluation process of the
RCM images and – as in OCT imaging – at present there is the
lack of specific staining possibilities for the recognition of certain
cell types comparable to immunohistochemistry. In this context
the use of fluorescence dye might enhance tissue recognition in
the future.
Although innovative HD-OCT shows a certain potential for
ex vivo evaluation of skin tumours, further technical developments
are still necessary here.
Acknowledgements
This work was supported by the Curd-Bohnewand-Fonds of
the University of Munich (to T.M.), by the Matthias Lackas
JEADV 2012
5
Foundation and the Dr. Helmut Legerlotz Foundation (to
C.B.). We thank Silke Krug for excellent technical assistance.
References
1 Gilbody JS, Aitken J, Green A. What causes basal cell carcinoma to be
the commonest cancer? Aust J Public Health 1994; 18: 218–221.
2 Flohil SC, de Vries E, Neumann HA, Coebergh JW, Nijsten T. Incidence, prevalence and future trends of primary basal cell carcinoma in
the Netherlands. Acta Derm Venereol 2011; 91: 24–30.
3 Bath-Hextall F, Leonardi-Bee J, Smith C, Meal A, Hubbard R. Trends
in incidence of skin basal cell carcinoma. Additional evidence from a
UK primary care database study. Int J Cancer 2007; 121: 2105–2108.
4 Bakos RM, Kriz M, Muhlstadt M, Kunte C, Ruzicka T, Berking C. Risk
factors for early-onset basal cell carcinoma in a German institution. Eur
J Dermatol 2011; 21: 705–709.
5 Sterry WGC, Korting HC. EDF Guidelines for Dermatology in Europe.
ABW Wissenschaftsverlag Berlin, Germany, 2009.
6 Loser C, Rompel R, Breuninger H et al. Microscopically controlled surgery (MCS). J Dtsch Dermatol Ges 2010; 8: 920–925.
7 Woerle B, Heckmann M, Konz B. Micrographic surgery of basal cell
carcinomas of the head. Recent Results Cancer Res 2002; 160: 219–224.
8 Breuninger H, Schaumburg-Lever G. Control of excisional margins by
conventional histopathological techniques in the treatment of skin
tumours. An alternative to Mohs’ technique. J Pathol 1988; 154:
167–171.
9 Boztepe G, Hohenleutner S, Landthaler M, Hohenleutner U. Munich
method of micrographic surgery for basal cell carcinomas: 5-year recurrence rates with life-table analysis. Acta Derm Venereol 2004; 84:
218–222.
10 Kunte C, Konz B. Current recommendations in the treatment of basal
cell carcinoma and squamous cell carcinoma of the skin. Hautarzt
2007; 58: 419–426.
11 Cunha D, Richardson T, Sheth N, Orchard G, Coleman A, Mallipeddi
R. Comparison of ex vivo optical coherence tomography with conventional frozen-section histology for visualizing basal cell carcinoma during Mohs micrographic surgery. Br J Dermatol 2011; 165: 576–580.
12 Maier T, Braun-Falco M, Hinz T, Schmid-Wendtner MH, Ruzicka T,
Berking C. Morphology of basal cell carcinoma in high definition optical coherence tomography: en-face and slice imaging mode, and comparison with histology. J Eur Acad Dermatol Venereol 2012 Apr 28. doi:
10.1111 ⁄ j.1468-3083.2012.04551.x. [Epub ahead of print]
13 Maier T, Braun-Falco M, Laubender RP, Ruzicka T, Berking C. Actinic
keratosis in the en-face and slice imaging mode of high definition optical coherence tomography and comparison with histology. Br J Dermatol 2012 Aug 4. doi: 10.1111 ⁄ j.1365-2133.2012.11202.x. [Epub ahead of
print]
14 Mogensen M, Thrane L, Jorgensen TM, Andersen PE, Jemec GB. OCT
imaging of skin cancer and other dermatological diseases. J Biophotonics
2009; 2: 442–451.
15 Mogensen M, Joergensen TM, Nurnberg BM et al. Assessment of optical coherence tomography imaging in the diagnosis of non-melanoma
skin cancer and benign lesions versus normal skin: observer-blinded
evaluation by dermatologists and pathologists. Dermatol Surg 2009; 35:
965–972.
16 Hinz T, Ehler LK, Hornung T, Voth H, Fortmeier I, Maier T et al.
Preoperative characterization of basal cell carcinoma comparing tumour
thickness measurement by optical coherence tomography, 20-mhz ultrasound and histopathology. Acta Derm Venereol 2012; 92: 132–137.
17 Gambichler T, Moussa G, Sand M, Sand D, Altmeyer P, Hoffmann K.
Applications of optical coherence tomography in dermatology. J Dermatol Sci 2005; 40: 85–94.
18 Gambichler T, Jaedicke V, Terras S. Optical coherence tomography in
dermatology: technical and clinical aspects. Arch Dermatol Res 2011;
303: 457–473.
ª 2012 The Authors
Journal of the European Academy of Dermatology and Venereology ª 2012 European Academy of Dermatology and Venereology
Maier et al.
6
19 Gambichler T, Orlikov A, Vasa R et al. In vivo optical coherence
tomography of basal cell carcinoma. J Dermatol Sci 2007; 45:
167–173.
20 Khandwala M, Penmetsa BR, Dey S, Schofield JB, Jones CA, Podoleanu
A. Imaging of periocular basal cell carcinoma using en face optical
coherence tomography: a pilot study. Br J Ophthalmol 2010; 94:
1332–1336.
21 Sauermann K, Gambichler T, Wilmert M et al. Investigation of basal
cell carcinoma [correction of carcionoma] by confocal laser scanning
microscopy in vivo. Skin Res Technol 2002; 8: 141–147.
22 Nori S, Rius-Diaz F, Cuevas J et al. Sensitivity and specificity of
reflectance-mode confocal microscopy for in vivo diagnosis of basal
cell carcinoma: a multicenter study. J Am Acad Dermatol 2004; 51:
923–930.
23 Gonzalez S, Tannous Z. Real-time, in vivo confocal reflectance microscopy of basal cell carcinoma. J Am Acad Dermatol 2002; 47: 869–874.
24 Chung VQ, Dwyer PJ, Nehal KS et al. Use of ex vivo confocal scanning
laser microscopy during Mohs surgery for nonmelanoma skin cancers.
Dermatol Surg 2004; 30: 1470–1478.
JEADV 2012
25 Patel YG, Nehal KS, Aranda I, Li Y, Halpern AC, Rajadhyaksha M.
Confocal reflectance mosaicing of basal cell carcinomas in Mohs surgical skin excisions. J Biomed Opt 2007; 12: 034027.
26 Gareau DS, Li Y, Huang B, Eastman Z, Nehal KS, Rajadhyaksha M.
Confocal mosaicing microscopy in Mohs skin excisions: feasibility of
rapid surgical pathology. J Biomed Opt 2008; 13: 054001.
27 Schule D, Breuninger H, Schippert W, Dietz K, Moehrle M. Confocal
laser scanning microscopy in micrographic surgery (three-dimensional
histology) of basal cell carcinomas. Br J Dermatol 2009; 161: 698–700.
28 González SGM, Halpern AC. Reflectance Confocal Microscopy of Cutaneous Tumors: An Atlas with Clinical, Dermoscopic and Histological Correlations. Informa Healthcare, London, 2008.
29 Hofman-Wellenhof RPG, Malvehy J, Soyer HP. Reflectance Confocal
Microscopy for Skin Diseases. Springer-Verlag, Berlin, Germany, 2012.
30 Maier T, Sattler EC, Braun-Falco M, Korting HC, Ruzicka T, Berking
C. Reflectance confocal microscopy in the diagnosis of partially and
completely amelanotic melanoma: report on seven cases. J Eur Acad
Dermatol Venereol 2012 Feb 10. doi: 10.1111 ⁄ j.1468-3083.2012.04465.x.
[Epub ahead of print]
ª 2012 The Authors
Journal of the European Academy of Dermatology and Venereology ª 2012 European Academy of Dermatology and Venereology