Optical coherence tomography versus punch biopsy for diagnosis of basal cell carcinoma: a multicentre, randomised, non-inferiority trial

Summary

Background

Punch biopsy is the gold standard for diagnosis and subtyping of basal cell carcinoma. The aim of this study was to assess whether use of optical coherence tomography (OCT), a non-invasive imaging tool, might avoid the need for biopsy.

Methods

In a multicentre, randomised, non-inferiority trial, patients (aged ≥18 years) with an indication for biopsy of a suspected basal cell carcinoma outside the H-zone (high-risk zone) of the face were randomly assigned (1:1) to receive either OCT or punch biopsy (regular care) via a web-based randomisation system. Patients were enrolled from three participating centres in the Netherlands: Maastricht University Medical Centre+, Catharina Hospital Eindhoven, and Zuyderland Medical Centre Heerlen. Stratification factors for randomisation were participating centre and the grade of clinical basal cell carcinoma suspicion (high vs low). The primary endpoint was the proportion of patients free from a recurrent or residual lesion (malignant or premalignant) 12 months after treatment. Modified intention-to-treat and per-protocol analyses were conducted, with a predefined non-inferiority margin of –10%. This trial is registered with ClinicalTrials.gov number, NCT03848078, and is complete.

Findings

Between Feb 25, 2019, and Sept 2, 2020, 598 patients were enrolled and randomly assigned to either the regular care group (n=299) or the OCT group (n=299). Data on the primary endpoint were available in 553 patients (n=268 in the regular care group, n=285 in the OCT group). After median follow-up of 12·7 months (IQR 11·2–14·1) in the OCT group and 12·6 months (10·8–14·3) in the regular care group, 253 (94%) of 268 patients in the OCT group and 266 (93%) of 285 patients in the regular care group were free from recurrent or residual lesions (malignant or pre-malignant) 12 months after treatment. According to our modified intention-to-treat analysis, the absolute difference (OCT vs regular care) was 1·07% (95% CI –2·93 to 5·06; one-sided p=0·30), with the lower limit of the 95% CI not exceeding the predefined non-inferiority margin of –10%. Per-protocol analyses led to proportions free from a residual or recurrent lesion (premalignant or malignant) of 95% (250 of 263) in the OCT group and 94% (262 of 278) in the regular care group, and an absolute difference of 0·81% (95% CI –2·98 to 4·60; one-sided p=0·34).

Interpretation

OCT-guided diagnosis and treatment of basal cell carcinoma is non-inferior to regular care punch biopsy. Implementation of OCT for diagnosis of basal cell carcinoma could reduce the number of consultations and invasive procedures.

Funding

The Netherlands Organization for Health Research and Development and Maurits en Anna de Kock Stichting.

Introduction

In White populations, one in five people will develop a basal cell carcinoma.1, 2 For diagnosis of lesions suspected to be basal cell carcinoma, guidelines recommend a punch biopsy to guide the decision on optimal treatment.3, 4 Histopathological diagnosis is important to distinguish between basal cell carcinoma and non-basal cell carcinoma lesions and to determine the histopathological subtype. For superficial basal cell carcinoma, topical therapy might be prescribed; however, for non-superficial basal cell carcinoma, the width of resection margins or an indication for Mohs' micrographic surgery is based on the subtype. Besides the inconvenience of a biopsy, awaiting histopathological examination causes treatment delay. Optical coherence tomography (OCT) has emerged as a promising non-invasive tool for basal cell carcinoma diagnosis, generating real-time, in-vivo, cross-sectional images of tissue microarchitecture with a depth of 1·0–1·5 mm.⁵ OCT is based on light interferometry: the interference of two optical beams reflected by tissue produces distinguishable shades in the black and white spectrum, which allows the identification of morphological basal cell carcinoma characteristics.⁶

OCT might avoid the need for biopsy if an OCT diagnosis of basal cell carcinoma and subtype can be made with high confidence.7, 8, 9 A treatment plan can be made immediately and a diagnostic biopsy would only be taken in case of doubt. With this strategy, it has been reported that a punch biopsy could be omitted in 30–40% of patients, with low risk of misclassification.7, 8, 9, 10 There is a small risk that non-basal cell carcinoma lesions are misdiagnosed as basal cell carcinoma or that nodular or aggressive basal cell carcinoma subtypes are underdiagnosed as superficial basal cell carcinoma by OCT.

Research in context

Evidence before this study

We searched PubMed, Cochrane databases, reference lists of papers on optical coherence tomography (OCT) and basal cell carcinoma, controlled-trials.com, ClinicalTrials.gov, and the UK NHS centre for reviews and dissemination on March 6, 2018, for articles published in English, with no date limits. We used the search terms “optical coherence tomography or OCT”, “basal cell carcinoma or BCC”, “specificity”, and “sensitivity”. Inclusion criteria were populations of patients with a skin lesion suspected to be basal cell carcinoma, histological assessment with a punch biopsy or excision used as gold standard, and that sensitivity and specificity estimates could be derived from the study. Five prospective cohort studies fulfilled these inclusion criteria and were judged on the basis of Quality Assessment of Diagnostic Accuracy Studies-2 criteria. None of these studies had a low risk of bias, mostly due to the absence of transparency concerning patient flow and methods for the estimation of sensitivity and specificity. The reference standard was judged as unclear in all studies because the gold standard was not defined clearly or because it was not reported whether an independent, experienced dermatopathologist assessed the histopathological slides. Our review showed that the literature demonstrates promising results regarding the use OCT-guided diagnosis of basal cell carcinoma to justify dermatologists' interest in this technique. In December, 2018, a Cochrane systematic review on the use of OCT for diagnosing skin cancer concluded that conventional OCT might have a role in the diagnosis of basal cell carcinoma in clinically challenging lesions. The meta-analysis showed a higher sensitivity and specificity for OCT than for visual inspection and dermoscopy; however, due to a small number of studies and varying methodological quality, implications to guide clinical practice could not be drawn yet. Appropriately designed prospective comparative studies are needed.

Added value of this study

To our knowledge, this study is the only clinical trial so far that evaluates whether OCT-guided diagnosis and treatment of clinically suspected basal cell carcinoma is non-inferior to punch biopsy (ie, regular care) in terms of clinical effectiveness and cost-effectiveness.

Implications of all the available evidence

Our findings and the evidence generated justify OCT being considered for inclusion in international guidelines for basal cell carcinoma diagnosis. Implementation of OCT requires the reorganisation of current clinical practices, wherein a punch biopsy with 1 week waiting time for the results can be replaced by a one-stop-shop approach in around two-thirds of suspected basal cell carcinoma cases. An important condition for the successful implementation of OCT in clinical practice is sufficient training. It is crucial that criteria be set for adequate diagnostic performance and for the time and training required to achieve good diagnostic performance.

To date, it remains unclear to what extent misclassifications would result in a higher risk of treatment failure. We, therefore, conducted a randomised, controlled trial with the aim of ruling out that OCT-guided diagnosis and treatment results in an unacceptable increase in treatment failures when compared with regular care.