Abstract
A mainstay treatment of head and neck basal cell carcinoma (BCC) is excision with a fixed surgical margin. Large-scale evaluation of post-surgical margin status is necessary to improve outcomes. This cross-sectional study assessed margin outcomes following excision of head and neck BCCs across Danish health-care settings, including risk factors for tumour-positive margins. Histopathology records on BCCs excised by all medical specialities between 1 January and 31 December 2022 were retrieved from the Danish Patho-logy Bank. Cases with intraoperative microscopic margin assessment were excluded. Descriptive analyses and multivariable logistic regression determined complete excision proportions and risk factors for positive margins. Including 9,291 BCC excisions in 7,741 patients, the tumour-free proportion was 80.0% (95% CI: 79.2–80.8%). Significant variation occurred by histopathological subtype, with tumour-free margins ranging from 60.6% in morpheaform to 83.6% in nodular BCCs. Multivariable analyses also identified morpheaform, micronodular, infiltrating, and superficial subtypes, auricular/postauricular locations, small specimen size, and advancing age as risk factors for tumour-positive margins. The nose was a risk factor in univariable analysis. Simple excision of head and neck BCCs showed notable variation in margin outcomes by patient, tumour, and specimen profile. To minimise recurrence and optimise resources, increased awareness of positive margins is essential.
Key words: basal cell carcinoma, margins of excision, skin neoplasms, operative surgical procedures, keratinocytes
SIGNIFICANCE
A mainstay treatment for basal cell carcinomas of the head and neck is simple excision with a fixed surgical margin. This large-scale national cross-sectional study assessed 9,291 head and neck basal cell carcinoma excisions throughout 2022. Overall, 4 out of 5 basal cell carcinomas (80.0%; 95% confidence interval 79.2–80.8%) were excised with free margins, with the highest odds of tumour-positive margins in high-risk histopathologic subtypes, superficial basal cell carcinomas, auricular, postauricular, nasal anatomic locations, and small specimen sizes.
Basal cell carcinoma (BCC) is one of the most common cancers in fair-skinned populations, with ultraviolet exposure comprising the major extrinsic aetiological driver (1, 2). Rising BCC incidence has been observed globally, partly due to ageing populations, deleterious sun behaviours, and improved disease awareness and detection (3, 4). Histopathologically, BCCs are categorized into subtypes, including the frequently occurring and less aggressive superficial and nodular tumours, as well as the more rare and aggressive micronodular, infiltrative, morpheaform, and basosquamous subtypes (5). Although BCCs rarely metastasize, the risk of tumour recurrence and development of subsequent lesions is substantial. Considerable morbidity and resource consumption are associated with BCC management (6, 7).
BCC treatment is tailored to suit the patient profile, specific tumour subtype and size, anatomic location, and local resource constraints (8). For BCCs of the head and neck, surgical excision under local anaesthesia is considered standard treatment, typically using surgical margins of 3–5 mm, sometimes up to 10 mm (9). Achiev-ing free margins is important: positive surgical margin has been found to be the strongest predictor of recurrence after simple excision of head and neck BCCs (odds ratio [OR] 6.71; 95% confidence interval [CI] 3.69–12.2) (10). The European treatment guideline report recurrence rates from 26% to 41% for incompletely excised tumours, with up to 5 years’ follow-up (8). In the head, achieving free margins is key to avoiding additional procedures, preserving the function and aesthetics of facial structures. Ensuring free surgical margins has implications for healthcare systems, lowering the need for intensified follow-up and additional interventions (11).
Simple excision involves demarcating the tumour’s borders and excising it with a margin of healthy tissue, typically including the subcutaneous plane for BCCs. Most procedures are done under local anaesthesia, with few exceptions. Intraoperative histopathological assessment excludes the procedure from being classified as a simple excision.
To minimise the BCC burden and optimise healthcare resource utilization, current treatment standards require continued evaluation. This large-scale study examined the proportion of simple excisions of BCCs of the head and neck that achieve tumour-free margins across all healthcare sectors in Denmark over the course of a year. Possible risk factors for tumour-positive margins, including patient, tumour, and specimen characteristics, were secondarily examined.
MATERIALS AND METHODS
Study design, data source, and review board approval
The study was a cross-sectional analysis of data from the Danish Pathology Data Bank (Patobank) originating from the period 1 January 2022–31 December 2022. The database contains detailed, nationwide records of all pathology specimens analysed across Denmark irrespective of primary, secondary, or tertiary provider origin. Data from Patobank is of high quality and completeness, offering detailed information on histopathological specimens (12). Demographic data registered in the database include patients’ personal identification numbers (civil registration number [CPR]), age, and sex. Data related to BCC excisions include a unique record ID, ICD-10 codes, histopathological diagnoses, tumour subtypes, anatomic locations, procedure codes, and free text macro- and microscopic specimen descriptions including dimensions.
Ethical approval
The study was approved by the governing body of data privacy protection, Research Law, Capital Region of Denmark, with approval number p-2023-14865 and by the governing body of medical records access, Team for Records Data, Science Ethics and Research Support, Capital Region of Denmark, with approval number R-23059658. According to Danish legislation, register-based studies are not required to gain acceptance by the national ethics committee. The study was performed in accordance with the Declaration of Helsinki.
Data retrieval and variable definitions
Described in detail in Appendix S1, data were retrieved by applying the Danish version of the Systematized Nomenclature of Medicine Clinical Terms (SNOMED CT) in a combined search string (13). Extracted anatomic locations were aggregated into the following groups: corners of the eye, eyelids, and periorbital areas were categorized as periorbital areas; cheeks, mandibular borders, and preauricular areas as the cheek; and neck and throat as the neck. Data on histopathological BCC subtype were extracted first according to the SNOMED CT, or, if unavailable, from free text derived from macro- and microscopic specimen descriptions. Data cleaning was completed using a set of algorithms described in detail in Appendix S1, and Tables SI and SII. For specimens where a combination of two or more histopathological subtypes was described, the most aggressive subtype was noted in the following descending order: basosquamous, morpheaform, infiltrative, micronodular, nodular, or superficial.
Specimen resection size, based on the largest noted specimen dimension in mm, was only available as free text in the macroscopic description of pathology records and was extracted using text search algorithms. To minimise the impact of erroneous records, all conspicuously large specimen sizes (50–499 mm) were manually reviewed and removed if obvious errors were identified. Due to the high likelihood of error for specimen sizes > 500 mm, these records were altered to “not available” regarding specimen size. The cutoff was chosen based on feasibility considerations for manual review, rather than specific clinical criteria. Categorization of surgical specimen resection sizes was decided by inspection of frequency distribution diagrams, sample sizes in each category to provide statistical strength, and consensus in the author group. Of note, macroscopic tumour size and surgical margin dimension were not available in Patobank, and thus not included in the analysis. Macroscopic tumour details regarding tumour borders, i.e., if tumours were well demarcated or not, were also not available in the database.
Inclusion and exclusion criteria
The study included all BCCs (SNOMED CT codes M809xx) located on the head or neck (codes T015xx, T021xx-T023xx, T21xxx, TXxxxx, or TY01xx-TY06xx) performed in primary, secondary, or tertiary healthcare settings irrespective of treatment provider. Excluded from the study were BCCs coded as biopsies, punch biopsies, incisional biopsies, curettages, intraoperative frozen sections, repeat sections, or recurrences (codes P30610, P30613, P3061A, P30626, P32200, P30624, or ÆYYY07) (Fig. 1). The study did not assess tumour size, surgical margin dimensions, applications of methods to demarcate the tumour perioperatively (e.g., dermoscopy or intraoperative microscopic margin control), skin closure techniques, possible complications to treatment, or recurrence rate.
Fig. 1.
Data inclusion flowchart. Adapted from Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021; 372: n71. https://doi.org/1136/bmj.n71. *One histopathology report can contain descriptions of several specimens (e.g., when a clinician surgically excises a tumour and performs a biopsy on another tumour on the same patient in 1 setting). Therefore, reports containing more than 1 specimen description had to be split to single specimen results as preparation for further analysis. ‡Similarly, searching for data in Patobank, their system outputs any histopathology record fulfilling the inclusion and exclusion criteria, even when a particular specimen description in the report does not meet the criteria. Consequently, after splitting records into single specimen results, another round of filtering for inclusion and/or exclusion criteria had to be carried out.
Information on the context of excisions, including whether the procedure was intended to be curative or diagnostic, was not available. Regarding surgical specimens, as extremely small sizes carried a high risk of being curettage/biopsy samples that had escaped our search filter rather than surgical excisions, we chose to exclude all excisions with sizes <5 mm. Sizes of 5–6 mm were manually reviewed and corrected if obvious errors were identified (Fig. 1).
Study outcomes
The proportion of complete excisions, based on tumour-free margins reported in retrieved pathology records (SNOMED CT free margins M09400), was calculated as a percentage (%) of all surgical excisions of head and neck BCCs. Data on remaining excisions comprised records describing tumour-positive margins (not free margins M09401) or margin status unavailable (margins not assessable M09402).
To evaluate risk factors for tumour-positive margins, uni- and multivariable logistic models incorporated and adjusted for patient, tumour, and specimen variables, including patient age, sex, anatomical tumour localization, histopathologic BCC subtype, and surgical specimen resection size.
Statistics
Categorical values are presented as a frequency with proportions, while age is presented as medians and interquartile ranges (IQR). Complete excision proportions with 95% CIs are calculated using an asymptotic normal approximation. The odds ratios (ORs) and adjusted ORs were obtained through uni- and multivariable logistic regression, respectively, using the statsmodel package version 0.14.1 in Python utilizing the Newton-Raphson method. Data cleaning and statistics were performed in Python version 3.7.4. (Python Software Foundation; https://www.python.org/).
RESULTS
Patient and tumour characteristics
A total of 9,964 BCCs of the head and neck treated with simple excision were included. At time of surgery, median patient age was 76 years with a female:male ratio of 1:1 in 8,378 unique patients. Information on surgical margin status was unavailable for 673 excisions (6.75%), prompting exclusion from subsequent analysis (Table I).
Table I.
Patient, tumour, and specimen characteristics stratified by tumour subtype
| Characteristic* | Totalb | BCC subtypea | ||||||
|---|---|---|---|---|---|---|---|---|
| Superficial | Nodular | Micronodular | Infiltrating | Morpheaform | Basosquamous | Not available | ||
| Patient and material | ||||||||
| Number of unique patients | 8,378 | 603 | 5,942 | 160 | 1,142 | 70 | 36 | 376 |
| Percentage of females/males | 49.6/50.4 | 58.2/41.8 | 49.1/50.9 | 59.4/40.6 | 46.8/53.2 | 47.1/52.9 | 44.4/55.6 | 45.2/54.8 |
| Median age at time of surgery in years (IQR) | 75.6 (66.9–81.4) | 75.2 (65.5–80.1) | 75.4 (66.8–81.2) | 74.7 (63.7–81.5) | 77.1 (69.6–82.8) | 76.9 (68.2–82.3) | 79.5 (75.3–83.2) | 76.8 (69.0–82.2) |
| Number of excisions (% of total) | 9,964 | 659 (6.6) | 6,743 (67.7) | 165 (1.7) | 1,217 (12.2) | 71 (0.7) | 36 (0.4) | 397 (4.0) |
| Anatomic location | ||||||||
| Scalp | 985 (9.9) | 104 (15.8) | 678 (10.1) | 11 (6.7) | 96 (7.9) | 8 (11.3) | – | 36 (9.1) |
| Forehead | 1,475 (14.8) | 113 (17.1) | 1,011 (15.0) | 29 (17.6) | 185 (15.2) | 11 (15.5) | – | 52 (13.1) |
| Temples | 1,134 (11.4) | 93 (14.1) | 747 (11.1) | 20 (12.1) | 155 (12.7) | 8 (11.3) | – | 58 (14.6) |
| Postauricular areas | 288 (2.9) | 13 (2.0) | 198 (2.9) | – | 35 (2.9) | – | – | 20 (5.0) |
| Ears | 820 (8.2) | 26 (3.9) | 476 (7.1) | – | 186 (15.3) | 17 (23.9) | 6 (16.7) | 42 (10.6) |
| Periorbital areas | 659 (6.6) | 22 (3.3) | 423 (6.3) | 14 (8.5) | 52 (4.3) | – | – | 23 (5.8) |
| Nose | 1,207 (12.1) | 59 (9.0) | 824 (12.2) | 17 (10.3) | 134 (11.0) | 5 (7.0) | 6 (16.7) | 44 (11.1) |
| Cheeks | 1,712 (17.2) | 104 (15.8) | 1,201 (17.8) | 26 (15.8) | 214 (17.6) | 9 (12.7) | 5 (13.9) | 65 (16.4) |
| Lips | 310 (3.1) | 14 (2.1) | 232 (3.4) | 8 (4.8) | 30 (2.5) | – | – | – |
| Chin | 177 (1.8) | 8 (1.2) | 129 (1.9) | 8 (4.8) | 14 (1.2) | 0 (0.0) | – | – |
| Neck | 753 (7.6) | 88 (13.4) | 512 (7.6) | 13 (7.9) | 69 (5.7) | 5 (7.0) | – | 26 (6.5) |
| Other facial areas | 444 (4.5) | 15 (2.3) | 312 (4.6) | 11 (6.7) | 47 (3.9) | – | – | 22 (5.5) |
| Surgical specimen resection size | ||||||||
| > 3.5 cm | 543 (5.4) | 50 (7.6) | 324 (4.8) | – | 94 (7.7) | – | 5 (13.9) | 30 (7.6) |
| 2.5–3.5 cm | 1,657 (16.6) | 95 (14.4) | 1,074 (15.9) | 28 (17.0) | 288 (23.7) | 21 (29.6) | 9 (25.0) | 81 (20.4) |
| 1.5–2.5 cm | 4,278 (42.9) | 297 (45.1) | 2,950 (43.7) | 72 (43.6) | 547 (44.9) | 32 (45.1) | 16 (44.4) | 169 (42.6) |
| 0.5–1.5 cm | 3,325 (33.4) | 210 (31.9) | 2,290 (34.0) | 56 (33.9) | 266 (21.9) | 9 (12.7) | 6 (16.7) | 110 (27.7) |
| Not available | 161 (1.6) | 7 (1.1) | 105 (1.6) | – | 22 (1.8) | – | 0 (0.0) | 7 (1.8) |
All entries are of the format “n (percentage of number of excisions)” unless otherwise stated. Fields with “–” are omitted for privacy reasons due to n < 5 or distinguishable n.
n = 673 tumours (n = 637 patients) are not counted in subtypes due to missing or not assessable margin outcome. BCC–SCC mixed tumours not shown due to n < 5.
The total column summarizes all specimens included from the population, whereas the subtype columns count only specimens included for analysis.
IQR: interquartile range.
In the remaining 9,291 excisions, nodular BCC (nBCC) was the most common histopathological subtype (67.7%), followed by infiltrating (12.2%), superficial (6.6%), unknown (4.0%), micronodular (1.7%), morpheaform (0.7%), or basosquamous (0.4%). Due to the small sample of mixed BCC–SCCs, this histopathological subtype was omitted from further analysis. Based on anatomic location, tumours located on the cheek, forehead, nose, or temple accounted for more than half of included BCCs, with the remainder located on the scalp, ears, neck, periorbital areas, “other facial areas”, lips, postauricular areas, and chin. Surgical specimen resection sizes of 1.5–2.5 cm were the most frequent (42.9%), while > 3.5 cm were rare (5.4%) (Table I). Included surgical specimen resection sizes ranged from 0.5 cm to 5.04 cm (99th percentile).
Margin status following simple excision
Tumour-free margins were recorded in 7,436 out of 9,291 pathology reports, yielding a complete excision proportion of 80.0% (95% CI 79.2–80.8%) (Table II). Stratifying according to histopathologic subtypes, morpheaform BCCs had the lowest proportion of free margins at 60.6% (95% CI 49.2–71.9%), while basosquamous and nodular BCCs had the highest at 86.1% (95% CI 74.8–97.4%) and 83.6% (95% CI 82.7–84.5%), respectively. Tumour-free margin proportions were thus observed in the following increasing order: morpheaform, micronodular, infiltrating, superficial, nodular, and basosquamous (Table II).
Table II.
Excisions and margin status stratified by tumour subtype
| BCC subtype | Number of excisions with free margins, n | Total number of Excisions, n | Proportion of complete excisions, % | 95% CI |
|---|---|---|---|---|
| Overall | 7,436 | 9,291 | 80.0 | 79.2–80.8% |
| Superficial | 497 | 659 | 75.4 | 72.1–78.7% |
| Nodular | 5,635 | 6,743 | 83.6 | 82.7–84.5% |
| Micronodular | 106 | 165 | 64.2 | 56.9–71.6% |
| Infiltrating | 820 | 1,217 | 67.4 | 64.7–70.0% |
| Morpheaform | 43 | 71 | 60.6 | 49.2–71.9% |
| Basosquamous | 31 | 36 | 86.1 | 74.8–97.4% |
| Not available | 304 | 397 | 76.6 | 72.4–80.7% |
BCC: basal cell carcinoma.
Risk factors for tumour-positive margins
Presented in Fig. 2, patient, specimen, and tumour factors were linked to increased odds of tumour-positive surgical margins in the univariable and multivariable analyses. Regarding patient characteristics, increasing age was weakly associated with a minimally increased multivariable odds ratio (mOR) for positive margins (for every 1-year increase: mOR 1.02; 95% CI 1.01–1.02), while no significant differences was detected between the sexes (Fig. 2).
Fig. 2.
Uni- and multivariable odds ratios (ORs) of tumour-positive margins based on patient, specimen, and tumour characteristics. ORs are presented in numerically increasing order followed by not available (N/A) for sex, surgical specimen resection size, and subtype, while ORs for anatomic location are presented in topographic cranial to caudal order, followed by “Other facial areas”.
Stratifying by histopathological subtype, superficial, infiltrating, micronodular, and morpheaform BCCs were associated with significantly increased mORs for positive margins as compared twith nBCCs, with the highest mOR detected for morpheaform tumours (mOR 4.10; 95% CI 2.49–6.75) (Fig. 2). Overall, high-risk histopathologic subtypes were associated with 2.4-fold higher odds of tumour-positive margins vs nBCCs (mOR 2.42; 95% CI 2.13–2.75). More specifically, morpheaform and micronodular BCCs carried a greater than 4- and 3-fold higher odds of positive margins compared with nBCCs, respectively. Superficial BCCs (sBCC) also had 81% higher odds of tumour-positive margins vs nBCC.
Based on anatomic location and using the scalp as a reference, BCCs located on the ears and postauricular areas had significantly increased mOR for positive surgical margins (mOR 1.44; 95% CI 1.13–1.84, and mOR 1.62; 95% CI 1.16–2.25). In univariable analysis, BCCs on the nose were also associated with a significantly increased OR for positive margins (OR 1.67; CI 1.35–2.07), but the association disappeared upon multivariable analysis (mOR 1.07; 95% CI 0.85–1.34). Conversely, tumours located on the periorbital area and cheek were associated with significantly lower mORs (mOR 0.75; 95% CI 0.56–0.99, and mOR 0.71; 95% CI 0.57–0.89) (Fig. 2).
Regarding surgical specimen resection size, small specimens < 1.5 cm were associated with the highest mOR for positive margins (mOR 2.80; 95% CI 2.19–3.58) using specimens > 3.5 cm as reference. Indeed, all specimen categories > 1.5 cm displayed lower mORs for positive margins compared with specimens < 1.5 cm in size (Fig. 2).
DISCUSSION
In one of the largest samples to date, this nationwide cross-sectional study assessed 9,291 head and neck BCC excisions from every level of the Danish healthcare sector throughout 2022 (14). Tumour-free margins were achieved in 80.0% of simple excisions, with notable variation in tumour-positive margins depending on histopathological BCC subtype, anatomic location, surgical specimen resection size, and age. High-risk histopathological subtypes, which represented 15% of included excisions, were associated with an over 2-fold higher odds of tumour-positive margins compared with nBCCs. Tumours on or behind the ears were similarly associated with 44–62% higher odds of positive margins vs scalp tumours. Small surgical specimen size < 1.5 cm was also associated with an elevated risk of positive margins. In sum, this study provides insight into the factors that impact margin status, an important surgical outcome for simple excisions of BCCs (14).
Generally, in BCC excision, the goal is complete tumour removal with free surgical margins, preserved function, and an acceptable aesthetic outcome. Minimizing morbidity may in some instances be more appropriate than a zero-tolerance for recurrence. A current British guideline suggests aiming for a ≥ 95% excision rate with free margins, while other European and American guidelines do not provide explicit targets (8, 15, 16). Our proportion of 80% falls short of the ≥ 95% target which, at face value, signals a need for more regular assessment of treatment outcomes. However, this study did not address the context of these excisions. Whether surgeries were intended to be curative, diagnostic, or palliative, was not known. Further, we analysed excisions irrespective of medical setting, ranging from general practices to tertiary centres. A Swedish study found that between 15.9% and 37.0% of high-risk BCCs of the head were incompletely excised (17).
Variable margin outcomes are reported in larger studies in the literature (Table III). Assessing all healthcare settings including primary, secondary, and tertiary sectors, Cho et al. reported free margins in 84.0% of 2,202 excisions of scalp BCCs (18). In a secondary sector study, a 78.0% free margin proportion was reported in 1,570 head and neck BCCs (19). Including 3,309 head and neck BCCs excised in a tertiary sector plastic surgery setting, an 84.2% free-margin proportion was noted, while an 86.9% proportion was reported for 2,092 BCC excised in a dermatology setting (20, 21). Other tertiary plastic surgery studies found considerably higher complete proportions up to 96.5% (Table III) (22). These studies are, however, heterogeneous and include different patient populations. Making direct comparisons between them is challenging.
Table 3.
Comparison of margin status with other studies
| Author | Year | Tumours, n | Proportion of complete excisions, % | Setting | Risk of bias |
|---|---|---|---|---|---|
| Betti et al. (21) | 2012 | 2,092 | 86.0 | Dermatology, tertiary | Moderate |
| Cho et al. (18) | 2016 | 2,202* | 84.0 | All sectors | Moderate |
| Codazzi et al. (20) | 2014 | 3,309 | 84.2 | Plastic surgery, tertiary | Low |
| Gualdi et al. (30) | 2015 | 2,845 | 92.9 | Dermatology, tertiary | Moderate |
| Hansen et al. (31) | 2009 | 2,872 | 90.2 | General practitioners (GPs) with a special interest, secondary | Low |
| Kyrgidis et al. (32) | 2010 | 1,480 | 90.8 | Maxillofacial surgery, tertiary | Low |
| Malik et al. (23) | 2010 | 1,589 | 84.6 | Plastic surgery, tertiary | Low |
| Ramdas et. Al. (19) | 2018 | 1,570 | 78.0 | GPs, dermatology, plastic surgery, secondary | Low |
| Shahshahani et al. (24) | 2011 | 1,394 | 87.7 | Dermatology, tertiary | Moderate |
| Sherry et al. (22) | 2010 | 2,500 | 96.5 | Plastic surgery, tertiary | Moderate |
Table developed with permission and background data from Nolan GS, Kiely AL, Totty JP, Wormald JCR, Wade RG, Arbyn M, et al. Incomplete surgical excision of keratinocyte skin cancers: a systematic review and meta-analysis. Br J Dermatol 2021; 184: 1033–1044. https://doi.org/10.1111/bjd.19660.
Study included only scalp BCCs.
The relationship between margin status and histopathological BCC subtype in our study corresponds with well-established evidence; high-risk histopathologic subtypes (including micronodular, infiltrative, and morpheaform BCCs) were significantly associated with incomplete tumour excision (18, 23, 24). Although superficial BCCs are considered low risk histopathologically, these tumours were also found to carry a significantly greater odds of positive margins compared with nBCCs in our study. This may reflect the greater challenge of delineating sBCCs as compared with nBCCs (25). Additionally, surgically excised sBCCs may belong to a clinically select subset of BCCs.
Certain anatomical locations, such as central areas of the face, are well-known risk factors for positive surgical margins (20). Subclinical spread has been linked to positive margin rates (26). Interestingly, we identified the nose to be a significant risk factor in the univariable analysis, while the association disappeared when adjusting for specimen size, histopathologic subtype, sex, and age. It may be interpreted that this possibly reflects the tendency for excisions in sensitive areas like the nose to have narrower margins due to limited tissue, as previously reported (22).
The inverse association between specimen size and incomplete excision was unexpected. The finding may reflect confounding, as surgical margin data, a crucial variable, was unavailable. Small specimens may indicate narrower margins. Consistent with this, Ramdas et al. found that specimen sizes ≤ 2 cm were linked to higher positive margin risk. Small specimen sizes may also result from limitations in surgical fields near vital facial structures. Contributing to the higher risk, small BCCs might be more difficult to clinically demarcate especially after biopsy or scar formation (19). Codazzi et al. recorded actual tumour sizes, noting that lesions < 0.5 cm were most associated with positive margins (20). Conversely, two studies of a total of over 2,000 head and neck BCCs found large tumour size to be a risk factor for positive margins (24, 25). Taken together, these studies seem to support that all facial BCCs should be considered high risk irrespective of size, as proposed by the National Comprehensive Cancer Network (NCCN) (27). The European Association of Dermato-Oncology (EADO) recommends wide surgical margins > 5 mm or micrographically controlled surgery for (i) nBCCs located on critical areas defined as the central face, eyelids, eyebrows, nose, lips, chin, ear, or periauricular areas; (ii) tumours with poorly defined margins or perineural invasion, and (iii) aggressive subtypes (8). Our findings imply that intraoperative margin-controlled surgery could be considered more often for high-risk BCC tumours of the head and neck.
This study has several weaknesses, including the data processing necessary to arrive at a final dataset. Some data disruption was unavoidable, e.g., biopsies erroneously registered as excisions. While excisions coded as recurrences were excluded, it is conceivable that a subset remained in our sample due to inadequate registry (28). Due to the large size and range of variables in the source data, we aggregated related variables, potentially overlooking discrete nuances on anatomic location. Most importantly, our study did not include two factors significantly associated with margin status: (i) clinical tumour size and (ii) surgical excision margin. Surgical excision margin is possibly the most significant independent factor ensuring complete tumour removal (8). Finally, we did not evaluate margin status’s effect on actual recurrence rate.
For simple excision, margin assessment is performed using 2D vertical sections of a portion of the specimen (in contrast to examination of the horizontal planes of the specimen and visualization of the whole tumour margin during 3D micrographically controlled surgery). As such, a segment of surgical excisions here deemed to have tumour-free margins could still recur up to a decade later (29). Finally, because our study combined data from all healthcare sectors, generalizability of our results is limited, and our outcomes may be poor reflections of those achieved in specific settings.
The main strength of this study was its population-based and large dataset of almost 10,000 excisions. Data were drawn directly from a national histopathology database with a high degree of completeness, providing records from primary-to-tertiary sectors, as well as public and private services in the Danish universal healthcare system. Extracting data from prospectively recorded patient data instead of, for example, via audit, minimised recall and selection bias. The study thus offers a look into real-world outcomes for a surgical treatment for head and neck BCC across multiple sectors, useful in guiding clinical decision-making regarding treatment.
Conclusion
Surgical excision of primary head and neck BCCs is associated with notable variation in tumour-positive margins depending on patient, tumour, and specimen characteristics. Minimizing recurrence and optimizing use of resources are important in BCC treatment. A heightened awareness of the risk of positive surgical margins following simple excision is needed for especially high-risk and superficial histopathologic subtypes, auricular, postauricular, and nasal locations, and small specimen resection sizes.
Supplementary Material
ACKNOWLEDGEMENTS
The authors would like to thank M. Winther-Jensen, MSc, PhD, and M. Villumsen, MSc, PhD, Section for Data, Biostatistics, and Pharmacoepidemiology, Centre for Clinical Research and Prevention, Copenhagen University Hospital – Frederiksberg, Copenhagen, Denmark, for advice on statistical methods.
Funding Statement
Funding The study was executed as part of the Skin Cancer Innovation Clinical Academic Group (SCIN CAG) of Greater Copenhagen Health Science Partners (GCHSP), and the Danish Research Centre for Skin Cancer, a public–private research partnership between Copenhagen University Hospital – Bispebjerg, Aalborg University Hospital, and the Private Hospital Mølholm.
Data availability statement
Due to the nature of the research collecting data from a patient registry, Danish health legislation does not provide support for sharing data.
Ethics statement
The study was approved by the governing body of data privacy protection, Research Law, Capital Region of Denmark with approval number p-2023-14865, and by the governing body of medical records access, Team for Records Data, Science Ethics, and Research Support, Capital Region of Denmark, with approval number R-23059658. According to Danish legislations, register-based studies are not required to gain acceptance by the national ethics committee. The study was performed in accordance with the Declaration of Helsinki.
Disclosure statements
The authors MKT, JS, KEK, NLM, and GS report no conflict of interest. EW and MH received speakers’ honoraria for Sanofi in the 2022 Non-Melanoma Skin Cancer Forum.
REFERENCES
- 1.Little MP, Lee T, Kimlin MG, Kitahara CM, Zhang R, Alexander BH, et al. Lifetime ambient UV radiation exposure and risk of basal cell carcinoma by anatomic site in a nationwide U.S. Cohort, 1983–2005. Cancer Epidemiol Biomark Prev Publ Am Assoc Cancer Res Cosponsored Am Soc Prev Oncol 2021; 30: 1932–1946. 10.1158/1055-9965.EPI-20-1815 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Global Burden of Disease 2019 Cancer Collaboration, Kocarnik JM, Compton K, Dean FE, Fu W, Gaw BL, et al. Cancer incidence, mortality, years of life lost, years lived with disability, and disability-adjusted life years for 29 cancer groups from 2010 to 2019: a systematic analysis for the Global Burden of Disease Study 2019. JAMA Oncol 2022; 8: 420–444. 10.1001/jamaoncol.2021.6987 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Kappelin J, Green AC, Ingvar Å, Ahnlide I, Nielsen K. Incidence and trends of basal cell carcinoma in Sweden: a population-based registry study. Br J Dermatol 2022; 186: 963–969. 10.1111/bjd.20964 [DOI] [PubMed] [Google Scholar]
- 4.Schreuder K, Hollestein L, Nijsten TEC, Wakkee M, Louwman MWJ. A nationwide study of the incidence and trends of first and multiple basal cell carcinomas in the Netherlands and prediction of future incidence. Br J Dermatol 2022; 186: 476–484. 10.1111/bjd.20871 [DOI] [PubMed] [Google Scholar]
- 5.Cameron MC, Lee E, Hibler BP, Barker CA, Mori S, Cordova M, et al. Basal cell carcinoma: Epidemiology; pathophysiology; clinical and histological subtypes; and disease associations. J Am Acad Dermatol 2019; 80: 303–317. 10.1016/j.jaad.2018.03.060 [DOI] [PubMed] [Google Scholar]
- 6.Chen JT, Kempton SJ, Rao VK. The economics of skin cancer: an analysis of medicare payment data. Plast Reconstr Surg Glob Open 2016; 4: e868. 10.1097/GOX.0000000000000826 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Duarte AF, Sousa-Pinto B, Freitas A, Delgado L, Costa-Pereira A, Correia O. Skin cancer healthcare impact: A nation-wide assessment of an administrative database. Cancer Epidemiol 2018; 56: 154–160. 10.1016/j.canep.2018.08.004 [DOI] [PubMed] [Google Scholar]
- 8.Peris K, Fargnoli MC, Kaufmann R, Arenberger P, Bastholt L, Seguin NB, et al. European consensus-based interdisciplinary guideline for diagnosis and treatment of basal cell carcinoma-update 2023. Eur J Cancer Oxf Engl 2023; 192: 113254. 10.1016/j.ejca.2023.113254 [DOI] [PubMed] [Google Scholar]
- 9.Hædersdal M, Mosegaard M, Lamberg A, Olesen AB, Jemec GBE, Lorentzen H, et al. Guidelines regarding treatment of basal cell carcinomas [Internet]. Danish Dermatological Society; 2019. [cited 2024 Feb 25]. Available from: https://dds.nu/wp-content/uploads/2019/03/DDS-Guidelines-BCC_Final_2019.02.26.pdf [Google Scholar]
- 10.Szewczyk M, Pazdrowski J, Pabiszczak M, Więckowska B, Dańczak-Pazdrowska A, Golusiński W. Local recurrence risk in head and neck basal cell carcinoma. Otolaryngol Pol 2022; 76: 1–5. 10.5604/01.3001.0015.8568 [DOI] [PubMed] [Google Scholar]
- 11.Ranganath B, Teixeira RM, Patel T, Garza R, Murphy RX. Clinical and financial implications of positive margins after nonmelanoma skin cancer resection: a longitudinal evaluation. Ann Plast Surg 2021; 87: 80–84. 10.1097/SAP.0000000000002566 [DOI] [PubMed] [Google Scholar]
- 12.Bjerregaard B, Larsen OB. The Danish Pathology Register. Scand J Public Health 2011; 39: 72–74. 10.1177/1403494810393563 [DOI] [PubMed] [Google Scholar]
- 13.Erichsen R, Lash TL, Hamilton-Dutoit SJ, Bjerregaard B, Vyberg M, Pedersen L. Existing data sources for clinical epidemiology: the Danish National Pathology Registry and Data Bank. Clin Epidemiol 2010; 2: 51–56. 10.2147/clep.s9908 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Nolan GS, Kiely AL, Totty JP, Wormald JCR, Wade RG, Arbyn M, et al. Incomplete surgical excision of keratinocyte skin cancers: a systematic review and meta-analysis. Br J Dermatol 2021; 184: 1033–1044. 10.1111/bjd.19660 [DOI] [PubMed] [Google Scholar]
- 15.Nasr I, McGrath EJ, Harwood CA, Botting J, Buckley P, Budny PG, et al. British Association of Dermatologists guidelines for the management of adults with basal cell carcinoma 2021. Br J Dermatol 2021; 185: 899–920. 10.1111/bjd.20524 [DOI] [PubMed] [Google Scholar]
- 16.Kim JYS, Kozlow JH, Mittal B, Moyer J, Olencki T, Rodgers P, et al. Guidelines of care for the management of basal cell carcinoma. J Am Acad Dermatol 2018; 78: 540–559. 10.1016/j.jaad.2017.10.006 [DOI] [PubMed] [Google Scholar]
- 17.Ceder H, Ekström A, Hadzic L, Paoli J. Clinicopathological factors associated with incomplete excision of high-risk basal cell carcinoma. Acta Derm Venereol 2021; 101: adv00496. 10.2340/00015555-3856 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Cho M, Lee J, James CL, Marshman G, Huilgol SC. Scalp basal cell carcinoma: review of 2,202 cases. Dermatol Surg Off Publ Am Soc Dermatol Surg Al 2016; 42: 834–841. 10.1097/DSS.0000000000000783 [DOI] [PubMed] [Google Scholar]
- 19.Ramdas K, van Lee C, Beck S, Bindels P, Noordhoek Hegt V, Pardo L, et al. Differences in rate of complete excision of basal cell carcinoma by dermatologists, plastic surgeons and general practitioners: a large cross-sectional study. Dermatol Basel Switz 2018; 234: 86–91. 10.1159/000490344 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Codazzi D, Van Der Velden J, Carminati M, Bruschi S, Bocchiotti MA, Di Serio C, et al. Positive compared with negative margins in a single-centre retrospective study on 3957 consecutive excisions of basal cell carcinomas. Associated risk factors and preferred surgical management. J Plast Surg Hand Surg 2014; 48: 38–43. 10.3109/2000656X.2013.800526 [DOI] [PubMed] [Google Scholar]
- 21.Betti R, Radaelli G, Crosti C, Ghiozzi S, Moneghini L, Menni S. Margin involvement and clinical pattern of basal cell carcinoma with mixed histology. J Eur Acad Dermatol Venereol 2012; 26: 483–487. 10.1111/j.1468-3083.2011.04104.x [DOI] [PubMed] [Google Scholar]
- 22.Sherry KR, Reid LA, Wilmshurst AD. A five year review of basal cell carcinoma excisions. J Plast Reconstr Aesthetic Surg JPRAS 2010; 63: 1485–1489. 10.1016/j.bjps.2009.09.007 [DOI] [PubMed] [Google Scholar]
- 23.Malik V, Goh KS, Leong S, Tan A, Downey D, O’Donovan D. Risk and outcome analysis of 1832 consecutively excised basal cell carcinomas in a tertiary referral plastic surgery unit. J Plast Reconstr Aesthetic Surg JPRAS 2010; 63: 2057–2063. 10.1016/j.bjps.2010.01.016 [DOI] [PubMed] [Google Scholar]
- 24.Shahshahani MM, Razzaghi M, Ehsani AH, Normohamadpour P, Naraghi Z, Kamyab K, et al. Incidence of incomplete excision in surgically treated basal cell carcinomas and identification of the related risk factors Iran J Dermatol 2011; 14: 1–5. [PubMed] [Google Scholar]
- 25.Su SY, Giorlando F, Ek EW, Dieu T. Incomplete excision of basal cell carcinoma: a prospective trial. Plast Reconstr Surg 2007; 120: 1240–1248. 10.1097/01.prs.0000279148.67766.e1 [DOI] [PubMed] [Google Scholar]
- 26.Walker P, Hill D. Surgical treatment of basal cell carcinomas using standard postoperative histological assessment. Australas J Dermatol 2006; 47: 1–12. 10.1111/j.1440-0960.2006.00216.x [DOI] [PubMed] [Google Scholar]
- 27.Schmults CD, Blitzblau R, Aasi SZ, Alam M, Amini A, Bibee K, et al. Basal Cell Skin Cancer, Version 2.2024, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Cancer Netw 2023; 21: 1181–1203. 10.6004/jnccn.2023.0056 [DOI] [PubMed] [Google Scholar]
- 28.Holm A-S, Nissen CV, Wulf HC. Basal cell carcinoma is as common as the sum of all other cancers: implications for treatment capacity. Acta Derm Venereol 2016; 96: 505–509. 10.2340/00015555-2282 [DOI] [PubMed] [Google Scholar]
- 29.Van Loo E, Mosterd K, Krekels GAM, Roozeboom MH, Ostertag JU, Dirksen CD, et al. Surgical excision versus Mohs’ micrographic surgery for basal cell carcinoma of the face: A randomised clinical trial with 10 year follow-up. Eur J Cancer 2014; 50: 3011–3020. 10.1016/j.ejca.2014.08.018 [DOI] [PubMed] [Google Scholar]
- 30.Gualdi G, Monari P, Crotti S, Damiani G, Facchetti F, Calzavara-Pinton P, et al. Matter of margins. J Eur Acad Dermatol Venereol 2015; 29: 255–261. 10.1111/jdv.12504 [DOI] [PubMed] [Google Scholar]
- 31.Hansen C, Wilkinson D, Hansen M, Soyer HP. Factors contributing to incomplete excision of nonmelanoma skin cancer by Australian general practitioners. Arch Dermatol 2009; 145: 1253–1260. 10.1001/archdermatol.2009.270 [DOI] [PubMed] [Google Scholar]
- 32.Kyrgidis A, Vahtsevanos K, Tzellos TG, Xirou P, Kitikidou K, Antoniades K, et al. Clinical, histological and demographic predictors for recurrence and second primary tumours of head and neck basal cell carcinoma. A 1062 patient-cohort study from a tertiary cancer referral hospital. Eur J Dermatol EJD 2010; 20: 276–282. 10.1684/ejd.2010.0903 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Due to the nature of the research collecting data from a patient registry, Danish health legislation does not provide support for sharing data.