Basal cell carcinoma (BCC) is the most common cancer diagnosed in white populations worldwide. The rising incidence of BCC is becoming a major worldwide public health concern [1,2]. Between 1973 and 2009, the European standardized rate quadrupled from 40 to 165 per 100,000 person-years for men and from 34 to 157 for women, most likely because of more intensive UV exposure [3]. This is supported by previous published epidemiological literature indicating that ultraviolet radiation is an important risk factor for BCC, with a significant increase among outdoor workers [4,5]. Despite the low mortality from BCC, multiple and recurring tumors confer a high morbidity and considerable burden for health care providers and health budgets. Although BCC does not seem to have a strong effect on patients’ quality of life, patients suffering from BCC are definitely interested in efficacy, low recurrence rates, and cosmetic outcomes of their treatment [6]. Meanwhile, resources available at hospitals have not increased proportionally, and therefore, optimizing the effectiveness of present treatment modalities in daily dermatologic practice is necessary [7].

Clinically, BCC are characterized by small, translucent, or pearly papules, with raised teleangiectatic edges [8]. Most BCC occur in sun-exposed skin of the head and neck areas [9,10]. Sensitivity and positive predictive value of the clinical diagnosis of BCC by dermatologists have been reported to be 95.4% and 85.9%, respectively [11]. However, dividing between BCC subtypes is not always possible upon clinical assessment. To date, histological analysis of punch biopsy remains the gold standard to confirm the clinical diagnosis of BCCs and divide between the following subtypes: superficial (sBCC), nodular (nBCC), micro nodular (mnBCC) and infiltrating (iBCC). Of those, nBCC and sBCC have a less aggressive growth pattern in comparison to mnBCC and iBCC. Additionally, mixed type BCC (mtBCC) can be defined as a combination of subtypes and is frequently composed of aggressive subtypes [12]. Surgical excision remains the standard of treatment, with Mohs micrographic surgery typically utilized for high-risk lesions [13]. Based upon the histological growth pattern, BCC are surgically removed with a margin of either 3 mm (nBCC and sBCC) or 5 mm (mnBCC and iBCC) in accordance with current Dutch guidelines.

The use of real-time in vivo reflectance confocal microscopy (RCM) has proven successful to noninvasively diagnose BCC. Various studies have demonstrated that RCM is safe and accurate (sensitivity and specificity) to diagnose BCC [14-18]. Reported sensitivity and specificity for RCM in diagnosing BCC range from 83%-100% and 79%-97%, respectively [19-25]. Furthermore, Peppelman et al and Longo et al recently reported on RCM features that might divide between nodular, micronodular, superficial, and infiltrative subtypes of BCC [24,26,27].

In 2012, van der Geer et al reported on the feasibility of a one-stop-shop (OSS) concept for the treatment of skin cancer patients [28]. One-stop-shop implies that on the day of the initial outpatient clinic consultation, diagnosis and treatment planning both take place. In their study, preoperative frozen section histology was used to confirm BCC diagnosis and subtype. The mean throughput time was 4 hours and 7 minutes, no complications were observed, and patient satisfaction was high [28]. Incorporating RCM as a noninvasive diagnostic tool in a BCC OSS concept for lesions suitable for conventional surgical excision might further reduce the time between clinical diagnosis and treatment, administrative workload, and costs.

The aim of our study is to assess the efficacy and safety of the OSS concept, using real-time in vivo RCM (Vivascope 1500; Lucid Technologies, Henrietta, NY, USA) as a diagnostic tool, prior to the surgical management of new primary BCC, of all subtypes, in the general population. We hypothesize that compared to current standards of care, the OSS concept using RCM will not result in a significant increase of incomplete surgical excision margins on the final pathology report of confirmed BCC lesions. It is further hypothesized that in this OSS concept, RCM will have acceptable diagnostic accuracy (sensitivity and specificity) for diagnosing BCC and dividing between subtypes, throughput time will not increase, and patient satisfaction will be higher for participating subjects.

Patients will be recruited from the outpatient clinics of the Department of Dermatology, Academic Medical Center, University of Amsterdam (AMC), and the Department of Dermatology, the Netherlands Cancer Institute-Antoni van Leeuwenhoek Hospital (AVL), second-line and third-line reference centers. Consecutive patients with clinically suspected new primary BCC will be prospectively enrolled and randomly assigned to either the experimental (RCM-OSS) or control (standard of care) group during times the study associates will be available. Clinical assessment will be performed by an experienced, board-certified dermatologist. Clinical and dermoscopy pictures of the BCC lesion will be taken by a medical photographer. Patients with multiple clinically suspected new primary BCC lesions will be included for only the lesion most suitable for conventional surgical treatment according to the following order: (1) chest, (2) extremities, and (3) head and neck area.

The inclusion criteria are the following:

The investigators will enrol subjects at both study locations (AMC and AVL). Included patients with clinically suspected new primary BCC lesions will be randomly allocated to the different diagnostic procedures. The investigators will obtain the patient’s consent. Each consecutive patient will be assigned a randomization number according to a computer-generated randomization list (ALEA) using random block sizes of 2, 4, 6, and 8 to ensure treatment concealment. Randomization will take place between the control and experimental group. This study will have an open label set-up. The patient and local investigator will not be blinded.

The randomization will be blinded. The pathologists analyzing the final excision specimen will be blinded to the patient’s history and to the results of RCM imaging. Whenever the histology of the punch biopsy is not required in the diagnostic process of the final excision specimen, the pathologist will also be blinded for those results. After initial RCM diagnosis by the study associates (DK and YE), two independent outcome assessors (M. Ulrich, Charite Berlin in Germany and C. Longo, Modena and Reggio Emilia in Italy) analyzing the RCM images will be blinded to the patient’s history and to the results of the final pathology report (reference standard).

We chose a cutoff of 95% as an acceptable radical BCC excision rate with standard of care based on our experience. Using the Miettinen and Nurminen confidence interval around the risk difference (24), with two groups of 38 patients, we will have 80% power to assess noninferiority of the OSS concept with RCM over usual care, considering an expected radical BCC excision rate of 95% in both arms, a noninferiority limit (delta) of 15%, and a one-sided alpha of 0.05. 

Incomplete surgical excision on the final pathology report of a routinely processed tissue specimen of confirmed BCC lesions (either by punch biopsy or RCM imaging) is the main outcome parameter. Assessment will be performed by an experienced board-certified pathologist. The number of incomplete excisions will be compared between the experimental and control group. Other assessments of included subjects with confirmed BCC lesions (either by punch biopsy or RCM imaging) will include the following:

BCCs will be divided into 5 main subtypes based on the histopathological growth pattern of the final excision specimen: superficial (sBCC), nodular (nBCC), micronodular (mnBCC), infiltrating (iBCC), and basosquamous (bBCC). In the case of mixed-type diagnosis, defined as two or more single growth patterns, the histology will be classified into single subgroups determined by the most aggressive component of the pathological feature according to the descending gradation from bBCC, iBCC, mnBCC, nBCC, to sBCC. The most aggressive component will determine the excision margin (5 mm versus 3 mm).

After obtaining written informed consent, the screening will be completed. Patients with clinically suspected new primary BCC lesions will be randomly allocated to the following regimes:

Data will be recorded on data entry forms and will be entered in a computer system for subsequent tabulation and statistical analysis. The data will be handled confidentially and anonymously. Furthermore, all information relevant to the treatment will be recorded in the electronic medical file.

All data will be collected and transferred to a Microsoft Excel database. The statistical analysis will be performed at the AMC using SPSS version 21.0. We will calculate the observed difference as the proportion of radical BCC excisions in the care-as-usual arm minus this proportion in the OSS with RCM arm, and calculate a one-sided 95% (or two-sided 90%) confidence interval for this difference using the Miettinen and Nurminen method [30]. The inferiority hypothesis will be rejected when the upper limit of this confidence interval does not exceed 15%. Side effects will be described per item.