Keywords
Carcinoma - Epithelial cells - Skin neoplasms - Keratinocytes
INTRODUCTION
Cutaneous squamous cell carcinoma (SCC) is a typical non-melanoma skin cancer occurring
in keratinocytes and the second most common skin cancer. It has a more invasive growth
pattern, higher recurrence rate, and higher potential to metastasize than basal cell
carcinoma (BCC) [[1]]. It develops as a result of exposure to sunlight and ultraviolet radiation, scars
caused by burns or injuries, chronic ulcer, extension of life expectancy, compromised
immune status, and genetic defects [[2]].
According to the Korea National Statistical Office, the average life expectancy of
South Koreans has increased from 78.6 years in 2005 to 81.94 years in 2015 and has
continued to rise [[3]]. Accordingly, the prevalence rates for SCC have also increased.
In principle, the treatment of SCC is complete resection of the primary tumor [[4]]. Thus far, clinicians have focused on examination of the horizontal extensibility
for determining the margin of complete resection; a safety margin has been emphasized
as a therapeutic strategy for treatment without recurrence. Therefore, although research
has been conducted on the tumor size and rate of metastasis to surrounding tissues
in cases of head and neck squamous cell carcinoma (HNSCC), no study on metastasis
according to the depth of tumor invasion has been reported.
The frequency of primary SCC has increased with population aging, and the recurrence
rate has also increased in accordance with extending survival time after primary resection.
Thus, there is a need for qualitative analysis of the depth of tumor resection by
analyzing the resection site and the depth of invasion according to the tumor size.
This study hypothesized that the horizontal size of HNSCC was proportional to its
vertical depth in the South Korean population. Establishing a relationship between
the horizontal and vertical safety margins of SCC through the confirmation of this
hypothesis could increase treatment completion.
METHODS
A prospective study was conducted on patients diagnosed with SCC who underwent surgical
treatment in the Department of Plastic and Reconstructive Surgery at our hospital
between November 2014 and October 2015. This study examined age, gender, presence
or absence of primary and recurrent tumors, presence or absence of complications,
and reconstruction methods.
Surgical methods
All surgeries were performed under general anesthesia. In patients with primary tumors,
the resection margin was designed, and complete resection was performed after setting
the safety margin to approximately 7 mm. The tumors were excised to a minimum 5-mm
depth from the tumor base by including the dermis completely, and they were resected
with inclusion of periosteum and cartilage. Specimens were obtained from the superior,
inferior, medial, and lateral resection margin and base after resection for frozen
section biopsy. When positive results were obtained, excision was performed to an
additional 5-mm depth and width until negative results were obtained ([Fig. 1A]).
Fig. 1 Surgical procedure
(A) Preoperative photo: designed excisional margin with safety margin. (B) Postoperative
photo: covered defect with rotation flap.
Regarding recurrent tumors, complete resection was performed by setting the safety
margin to approximately 10 mm. Additional resection was performed with the same method
used for primary SCC. The residual defects after complete removal of the tumors were
reconstructed using local flaps, including transposition flaps, island flaps, rotation
flaps, and advancement flaps, by maintaining facial aesthetic units as much as possible
([Fig. 1B]). All patients were discharged an average of 2.3 days after surgery, and all surgical
sutures were removed within 5 days after surgery.
Follow-up was conducted on an outpatient basis every 1, 2, 4, and 6 months and 1,
2, and 3 years after discharge. Macroscopic examination and computed tomography were
performed to examine recurrence and sequelae. The mean follow-up period was 8.9 months.
Pathological methods
The major and minor axis lengths of the surgically resected specimens were measured
using a microscopic ruler, and then fixed by formalin for 24 hours. Paraffin blocks
were prepared by serial sectioning of the fixed tissue at 2-mm intervals, followed
by the preparation of slides. Each slide was dewaxed using xylene; hematoxylin and
eosin staining was then applied. The stained slide was evaluated by the horizontal
size of SCC (major axis×minor axis, mm), the depth of tumor (mm), and invasion into
the surrounding tissues based on microscopic pathological findings ([Fig. 2]).
Fig. 2 Tumor measurements
a, horizontal safety margin; b, major axis of tumor; c, minor axis of tumor; d, invasion
depth. Large circle (○) and smaller circle (●) indicate excision margin and tumor
margin, respectively.
The major axis of the tumor, as identified by this study, represents the longest horizontal
length of the tumor, while the minor axis represents the longest horizontal length
perpendicular to the major axis. The size is defined as the product of the values
of the major axis and minor axis; the invasion depth is defined as the deepest vertical
length of the tumor measured by a microscope; and the invasion layer is defined as
the deepest tissue layer invaded by the tumor (e.g., dermis, subcutaneous tissue,
and muscle).
Statistical methods
Statistical analysis was performed with the SPSS software package, ver. 21.0 (IBM
Corp., Armonk, NY, USA). The statistical correlation between the major and minor axis
lengths of the tumor and the depth of tumor invasion (mm) was evaluated using the
Pearson correlation coefficient. The invasion was classified according to dermis,
subcutaneous layer, and muscle layer, and then evaluated by the correlation between
the major and minor axis lengths of the tumor. A P-value of <0.05 was considered a
statistically significant result.
RESULTS
Data were collected from 26 patients, with a mean age of 81.2 years, including 11
males and 15 females. The mean tumor size was 2.49×2.03 cm2. Eighteen patients had primary tumors and 8 patients had recurrent tumors after undergoing
treatment such as laser therapy or simple excision. All carcinomas were completely
removed after surgery, and there were no complications such as hematoma, seroma, partial,
or total flap necrosis after reconstructive surgery. There were no patients with recurrent
tumors during follow-up ([Table 1]).
Table 1
Patient characteristics
|
Characteristic
|
No. of patients (n = 26)
|
|
The values given in parentheses are the minimum and maximum values of the corresponding
item.
|
|
Mean age (yr)
|
81.2 (65–94)
|
|
Male:Female ratio
|
11:15
|
|
Mean major axis of tumor gross length (cm)
|
2.5 (0.6–5.7)
|
|
Mean minor axis of tumor gross length (cm)
|
2.0 (0.3–5.4)
|
|
Mean invasion microscopic depth (mm)
|
7.2 (2–25)
|
|
Invasion layer ratio (dermis:subcutaneous tissue:muscle)
|
18:5:3
|
|
Primary tumor:Recurred tumor ratio
|
18:8
|
|
Complication (hematoma, seroma, necrosis, etc.)
|
0
|
|
Recurrence
|
0
|
Correlation between major axis length of tumor and depth of tumor invasion
The relationship between the major axis length of the tumor and the depth of tumor
invasion was proportional, with a Pearson correlation coefficient of 0.747 (P<0.001),
and showed statistically significant results ([Table 2]). No difference was observed between primary and recurrent carcinomas.
Correlation between minor axis length of tumor and depth of tumor invasion
In a comparison of the relationship between the minor axis length of tumor and depth
of tumor invasion, the Pearson correlation coefficient of 0.773 (P<0.001) was proportional,
and showed statistically significant results ([Table 2]). No difference was observed between primary and recurrent carcinomas.
Table 2
Relation between tumor size an invasion depth and layer
|
Variable
|
Invasion depth
|
Invasion layer
|
|
Major axis of tumor
|
|
|
|
r
|
0.747
|
0.032
|
|
P-value
|
< 0.001
|
0.881
|
|
No.
|
26
|
26
|
|
Minor axis of tumor
|
|
|
|
r
|
0.773
|
–0.050
|
|
P-value
|
< 0.001
|
0.815
|
|
No.
|
24
|
24
|
Correlation between horizontal tumor size and layer of tumor invasion
According to the classification of invasion, more than half of specimens (18 patients)
were localized in the dermis ([Table 1]). The relationship between the major and minor axis lengths of the tumor and invasion
showed no statistically significant results ([Table 2]).
Linear regression analysis was performed using the proportion between the major and
minor axis lengths of the tumor and the depth of invasion ([Fig. 3], [4]).
Fig. 3 Scatter plots (major axis)
Scatter plots with regression line between tumor major axis length and tumor invasion
depth.
Fig. 4 Scatter plots (minor axis)
Scatter plots with regression line between tumor minor axis length and tumor invasion
depth.
Y (mm, depth of invasion)=2.805×X (cm, length of major axis)+0.118, r2=0.558 (1)
Y (mm, depth of invasion)=3.131×X (cm, length of minor axis)+0.704, r2=0.597 (2)
In a comparison of the difference between the values of Y (the measured depth of invasion)
obtained by substituting the major and minor axis lengths into value X of the 2 completed
equations, Eq. (2) with X=length of minor axis showed higher similarity.
Table 3
Factors associated with high risk for recurrence in squamous cell carcinoma
|
High risk tumor characteristics
|
High risk size/location combinations
|
|
Poorly defined borders
|
≥6 mm in “mask areas” of the face (central face, eyelids, eyebrows, nose, periorbital,
lips, chin, ear, pre/post auricular skin, mandible, temple), genitalia, hands/feet
|
|
≥10 mm in the scalp, forehead, cheeks, neck
|
|
Recurrent lesion
|
|
|
Immunosuppressed patient
|
|
|
Site of prior radiotherapy or chronic inflammation (marjolin’s ulcer)
|
|
|
Rapidly growing tumor
|
|
|
Neurologic symptoms
|
|
|
Certain pathologic characteristics (perineural or vascular involvement, poor differentiation,
adenoid, adenosquamous or desmoplastic subtypes, thickness ≥ 2 mm or clark level IV
or V)
|
|
DISCUSSION
Cutaneous SCC, a typical non-melanoma skin cancer with BCC, shows aggressive features,
with a high rate of metastasis at 12.5% [[5]]. It develops due to exposure to sunlight and ultraviolet radiation, scars caused
by burns or injuries, chronic ulcer, extension of life expectancy, compromised immune
status, and genetic defects [[2]]. Although the reported frequency is 2.6 to 2.9 per 100,000 population [[6]], it will gradually increase with extensions to life expectancy.
Surgical resection is the treatment of choice for SCC, and nonsurgical treatments
include cryotherapy, curettage and electrodesiccation, and radiation therapy. Nonsurgical
treatments, however, are applied to patients who have difficulty in undergoing surgery
due to a high failure rate in recurrent tumors or large tumors; these nonsurgical
treatments have a high recurrence rate [[7]
[8]].
Surgical resection in SCC is based on tumor size. Surgical resection is performed
to include the tumor and a surrounding safety margin. However, controversy remains
over what constitutes an adequate safety margin. Brodland and Zitelli [[4]] recommend 4 mm for a low-risk tumor and 6 mm for a high-risk tumor. Guidelines
from the National Comprehensive Cancer Network recommend 4–6 mm for low-risk tumors
and 10 mm for high-risk tumors [[9]]. Based on the surgeon's experience, our hospital has set 7 mm as the safety margin
for primary tumor cases and 10 mm as the safety margin for secondary tumor cases.
However, what is most important is that the excision have clean margins. Invasion
of tumor cells on the resection margin is confirmed by pathological examination. When
positive results are obtained, additional resection should be performed to check for
negative results [[4]
[9]]. The low- and high-risk groups are classified according to tumor size and location,
histological classification, tumor characteristics, and invasion of surrounding tissues
([Table 3]) [[9]].
When the depth of SCC reaches the subcutaneous tissues or beyond, the metastasis rate
increases dramatically [[10]
[11]]. Some doctors believe that tumor depth is the most important factor in predicting
metastasis of SCC [[11]]. Moreover, since the tumor relapse rate is higher at greater depth, it is important
to identify the depth of a tumor during treatment.
Five-year recurrence of cutaneous SCC varies depending on tumor size. It reaches approximately
8% and 15% with small (diameter <2 cm) and large (diameter >2 cm) tumors, respectively
[[12]
[13]
[14]]. Regarding recurrent tumors, the recurrence rate is double, and the rate of local
metastasis reaches 5% to 30% [[12]].
With the increase in life expectancy, it is expected that not only incidence of SCC
but also the recurrence rate will increase as the follow-up period after treatment
is extended. Thus, complete treatment of primary SCC is important.
Incomplete resection has been evaluated as the most common cause of an increase in
tumor recurrence and metastasis risk in the treatment of SCC [[12]
[15]]. In other words, there is a need for complete resection considering tumor extension
in order to lower the recurrence rate. To this end, a safety margin is important.
However, current standards of care for SCC specify only the horizontal safety margins.
Standards considering the depth of tumor invasion have yet to be established, and
no research has been conducted on the relationship between horizontal and vertical
extension of tumors and metastasis according to vertical extension of tumors. Therefore,
it is difficult to clarify whether recurrence due to incomplete resection is caused
by the remainder of the tumor on the horizontal resection margin or the remainder
of tissue on the tumor base.
In this study, serial section slides were prepared after surgical resection according
to existing guidelines for patients with SCC, and data were collected by pathological
examination of the base depth in each slide. With these data, the depth of the tumor
base was quantified by mapping based on the total horizontal area of the tumor. Statistical
analysis of the correlation between the major and minor axis lengths of the tumor
and the maximum depth of the tumor base was performed. A proportional relationship
between the major axis length of the tumor and the depth of tumor invasion was demonstrated
with a Pearson correlation coefficient of 0.747.
A proportional relationship of the minor axis length of the tumor and the depth of
tumor invasion was also demonstrated with a Pearson correlation coefficient of 0.773.
It can be concluded that horizontal growth and vertical growth of the cutaneous SCC
have a proportional relationship. This accords with our previously described hypothesis.
The following linear regression equations were obtained by performing a regression
analysis based on the horizontal tumor size (major and minor axis lengths) measured
using a microscope:
Y (mm, depth of invasion)=2.805×X (cm, length of major axis)+0.118, r2=0.558 (1)
Y (mm, depth of invasion)=3.131×X (cm, length of minor axis)+0.704, r2=0.597 (2)
For example, a 7.13-mm depth of invasion is obtained when substituting 2.5 cm as the
average measured length of the major axis of the patient into Eq. (1). A 6.97-mm depth
of invasion is obtained when substituting 2.0 cm as the average measured length of
the minor axis of the patient into Eq. (2), and this value is similar to 7.2 mm, the
measured average depth of invasion. Therefore, the use of this equation makes it possible
to estimate the depth of the tumor base by measuring the horizontal tumor size. This
means that a depth of at least 7.13 mm should be applied for complete resection for
patients with SCC measuring 2.5×2.0 cm. This also makes it possible to set a vertical
safety margin in proportion to tumor size.
A limitation of the present study is that the data were drawn from 26 patients, which
may not be sufficient for generalization with the formulae obtained. However, the
correlations between the data were statistically significant and the derived formulae
are expected to be validated when applied in future to other patients.
In order to perform complete resection of SCC, many plastic and reconstructive surgeons
and doctors have focused on the horizontal safety margin and surrounding tissues.
In this study, we clarified the proportional relationship between the horizontal size
and vertical depth of SCC and provided formulae for determining the vertical safety
margin. Performance of resection in consideration of the horizontal and vertical safety
margins of tumors based on these results could not only reduce risks such as tumor
recurrence by complete resection of the tumor base, but also minimize the defect size
of surgical sites by removing appropriate tissues.
