Efficacy of Photodynamic Diagnosis-Guided Mohs Micrographic Surgery in Primary Squamous Cell Carcinoma SU-YOUNG JEON, MD, KI-HO KIM, MD,
KI-HOON SONG, MD*
BACKGROUND Cutaneous squamous cell carcinoma (cSCC) usually has ill-defined margins because of its irregular invasive patterns. OBJECTIVE To evaluate the surgical efficacy of photodynamic diagnosis (PDD) in primary cSCC treated using Mohs micrographic surgery (MMS). METHODS & MATERIALS We examined 67 cases of biopsy-proven primary facial cSCC treated with MMS. The 67 SCC were divided into the two groups depending on PDD application: PDD group (n = 38, 56.7%) and non-PDD group (n = 29, 43.3%). We analyzed the differences in surgical features between the PDD and non-PDD groups. RESULTS The PDD group required fewer Mohs stages (1.37 vs 1.83, p = .02) and smaller surgical margins (8.03 vs 11.24 mm, p = .03). PDD showed additional benefits in terms of surgical margin and Mohs stage, especially in low-risk SCC, including thin (≤4 mm), small (≤20 mm), well-differentiated, and nonulcerative tumors (p < .05) but did not show beneficial effects in high-risk SCC (p > .05) during MMS. CONCLUSION PDD can increase surgical efficacy of primary cSCC during MMS. These benefits are more pronounced in low-risk SCC. We recommend PDD as a simple and useful technique for delineating the margins of low-risk SCC before MMS. The authors have indicated no significant interest with commercial supporters.
quamous cell carcinoma (SCC) is the second most common malignant tumor of the skin, and its incidence continues to increase every year.1 Although most SCC can be treated with surgery or other dermatological procedures, high-risk SCC is a challenge for dermatologic surgeons. Mohs micrographic surgery (MMS) is the most effective modality for removing cutaneous SCC (cSCC) because it conserves normal tissue, preserves function, and enhances cosmesis.2–4 In MMS, the initial determination of an appropriate tumor margin starts with an accurate assessment of clinical tumor extent. Because of high rates of subclinical spread in cSCC, visualization and palpation are
unreliable for determining tumor extent.4 An improved preoperative assessment of cancer margins may decrease the number of Mohs stage needed. Various methods for demarcating tumor margins during MMS have been introduced over the past few years.5–7 These techniques, including positron emission tomography,5 spectrophotometric intracutaneous analysis,6 and laser Doppler velocimetry,7 are highly complicated and require advanced training and specialized equipment. In contrast, photodynamic diagnosis (PDD) is a simple, noninvasive diagnostic technique in which a photosensitizer is applied to the lesions while the red fluorescence of selectively accumulated protoporphyrin IX (PpIX) is
*All authors are affiliated with the Department of Dermatology, Skin Cancer Center, College of Medicine, Dong-A University, Busan, Korea © 2013 by the American Society for Dermatologic Surgery, Inc. Published by Wiley Periodicals, Inc. ISSN: 1076-0512 Dermatol Surg 2013;39:1774–1783 DOI: 10.1111/dsu.12359 1774
JEON ET AL
observed using a Wood lamp to determine tumor margins.8,9 PDD was recently introduced to determine tumor margins before MMS because it is not complicated and does not require advanced training or specialized equipment. Studies have shown that the visible boundary observed using PDD corresponds closely to the histologic boundary and is thus helpful for determining the extent of surgery,8–13 but most of these studies evaluated the efficacy of PDD in basal cell carcinoma (BCC).8–12 In this study, we evaluated the surgical efficacy of PDD in patients with primary cSCC undergoing MMS.
Materials and Methods Study Design and Study Population Sixty-four consecutive patients with 67 biopsyproven primary facial cSCC lesions treated with MMS between January 2001 and December 2010 were enrolled in this study. Mean patient age was 71.1 (range 34–91). There were 23 men and 44 women (ratio 1:1.9). All patients had Fitzpatrick skin phototypes IV or V. The average disease duration was 37.6 months (range 1 month to 20 years). The institutional review board of Dong-A University Medical Center approved this retrospective, comparative, single-center study. The study protocol conformed to the guidelines of the 1975 Declaration of Helsinki. Methods The 67 SCCs were divided into two groups according to PDD application: PDD group (n = 38, 56.7%) and non-PDD group (n = 29, 43.3%). When the tumors had ill-defined margins under visual inspection, PDD was performed before surgery to delineate the tumor margins (Figures 1A and 2A). PDD-Guided MMS Using Aminolevulinic Acid or Methyl Aminolevulinic Acid Patients underwent PDD approximately 3 to 6 hours (methyl aminolevulinic acid (MAL),
3 hours; aminolevulinic acid (ALA), 6 hours) before MMS to delineate the tumor margins. Before the application of a photosensitizer, excessive scales and crusts were gently scraped off without bleeding and the lesions were cleansed using physiological saline gauze. A 1-mm layer of cream containing 16% MAL (Metvix; PhotoCure ASA, Oslo, Norway) or 20% ALA was then applied over the tumor lesions, including a 0.5- to 1-cm area of surrounding skin. The area was covered with polyurethane film (Tegaderm, 3M, Minneapolis, MN) and aluminum foil to prevent exposure to light. The two photosensitizers were randomly applied to the lesions (ALA, n = 19; MAL, n = 19). After incubation times of 3 to 6 hour (MAL, 3 hours; ALA, 6 hours), the occlusive dressing was removed, and the lesion was cleansed with saline to remove excess photosensitizer cream. To determine the margin of primary cSCC for MMS, the red fluorescence was marked with gentian violet under a Wood lamp (Ultraviolet Examination Light, model 31602, 356 nm; Burton Medical Products Corp., Chatsworth, CA, USA) (Figures 1 and 2). In the non-PDD group, tumor margins were determined and marked under direct visual observation. Data Collection All data were collected from medical records, including clinical photographs, pathology slides, and MMS operation sheets. High-risk SCC were identified according to the following standard criteria: invasive tumors of the ear and lip, tumors more than 4 mm deep, tumors more than 20 mm in diameter, ulcerative tumors, and moderately or poorly differentiated tumors.2,3,14,15 For the identification of high- or low-risk SCC, the clinicopathologic data, including tumor size, depth, and location; presence or absence of ulceration; and histologic subtypes, were evaluated. Surgical data included surgical margin, surgical depth, and number of Mohs stages required to clear the tumor. Pathology slides were obtained from the archives of the Department of Dermatology and Pathology,
PHOTODYNAMIC DIAGNOSIS IN SCC
Figure 1. Fluorescence and clinical images of a low-risk squamous cell carcinoma (SCC) with ill-defined margins. (A) Preoperative visual tumor size (tumor size 20 mm in diameter). (B) Biopsy specimen from the lesion showed a welldifferentiated SCC 3.5 mm deep (hematoxylin-eosin stain; original magnification: 940). (C) Fluorescence image after photodynamic diagnosis application (tumor size 28 mm in diameter, red line; final surgical margin, 32 mm in diameter, yellow line). (D) Final defect after one stage of Mohs micrographic surgery (defect size 32 mm in diameter). (E) The defect was covered with a transposition flap.
Dong-A University Medical Center, and two dermatologists who were blinded to the clinical information performed the histologic examination. Histologic SCC types were classified into well, moderately, and poorly differentiated according to the proportion of atypical squamous cells and the degree of keratinization. Tumor size was graded according to the greatest lesion diameter and measured with the naked eye in the non-PDD group, but in the PDD group, tumor size was based on the largest diameter of the red fluorescence (PDD margin) and measured under a Wood lamp after PDD application. Post-MMS defect size was based on the greatest diameter of the postoperative defect and measured after MMS removal. Surgical margin was designed and calculated by subtracting the value of the greatest diameter of each tumor from the greatest diameter
of the postoperative defect. Surgical depth was classified into categories of subcutaneous, muscle fascia or muscular, and periosteum or bony layer.
Statistical Analysis We categorized patients into the PDD group and nonPDD group according to PDD application. First, we analyzed the differences in surgical features between the PDD group and the non-PDD group in primary cSCC. Then we categorized the primary cSCC into a high-risk SCC group and a low-risk SCC group based on the following standard criteria: tumor location, size, and depth; presence or absence of ulceration; and histologic patterns. To compare the efficacy of PDD in high- and low-risk SCC treated with MMS, we performed a comparative analysis of the surgical margin and Mohs stage depending on PDD application. To investigate whether photosensitizer type
JEON ET AL
Figure 2. Fluorescence and clinical images of a high-risk squamous cell carcinoma (SCC) with ill-defined margins. (A) Preoperative visual tumor (tumor size 23 mm in diameter). (B) Biopsy specimen from the lesion showing a poorly differentiated SCC measuring more than 6 mm deep (hematoxylin-eosin stain; original magnification: 940). (C) Fluorescence image after application of photodynamic diagnosis (tumor size 25 mm in diameter, red line; final surgical margin 35 mm in diameter, yellow line). (D) Final defect after three stages of Mohs micrographic surgery (defect size 35 mm in diameter). (E) The defect was covered with a cheek transposition flap.
(MAL vs ALA) influenced surgical outcomes, we divided the PDD group into the ALA-PDD group and MAL-PDD group and analyzed the differences in surgical features between the two groups. All statistical analyses were performed using SPSS software (version 18.00; SPSS, Inc., Chicago, IL, USA). The categorical variables were analyzed using the Pearson chi-square test and Fisher exact test, and the continuous variables were analyzed using one-way analysis of variance and the Student t-test. p < .05 was considered statistically significant.
results of comparative analysis showed several significant differences (Table 1). After MMS, the mean number of Mohs stages required for complete tumor removal was lower in the PDD group (1.37 0.75) than in the non-PDD group (1.83 0.89) (p = .02). In the PDD group, 29 cases (76.3%) required one Mohs stage for complete tumor removal, whereas nine (23.7%) required two or more stages. In the non-PDD group, 13 cases (44.8%) required one Mohs stage for complete tumor excision, whereas 16 cases (55.2%) required two or more stages (p = .008).
Results Comparisons of Surgical Features Between the Non-PDD and PDD Groups We analyzed the differences in surgical features between the PDD and non-PDD groups, and the
Surgical margins were smaller for patients in the PPD group (8.03 5.02 mm) than for patients in the non-PDD group (11.24 6.83 mm) (p = .03). In the PDD group, 16 cases (42.1%) had a surgical margin of .99
SD, standard deviation.
three cases (10.3%) had a surgical margin of .05). Comparisons of Surgical Margin and Mohs Stage of High- and Low-risk SCC According to PDD Application We categorized the primary cSCC as high- and lowrisk SCC according to the previously described standard criteria and analyzed the differences in surgical margin and number of Mohs stages between high- and low-risk SCC in the PDD and non-PDD groups. In low-risk SCC, tumors measuring .05).
Discussion MMS is an effective surgical treatment for invasive skin cancers that theoretically facilitates complete resection. In numerous studies, MMS has been shown to have lower initial and 5-year recurrence rates and better cosmetic outcomes than other treatment modalities.2–4 More-accurate detection of tumor margins might decrease the number of Mohs stages and facilitate MMS. Some noninvasive modalities have been introduced to detect skin cancer tumor margins.5–9 PDD, a diagnostic technique based on fluorescence, can be used as a preoperative diagnostic technique that can help detect occult tumor borders of ill-defined skin cancers before MMS.10–12 Several recent reports have demonstrated excellent correlation between the fluorescence pattern of PDD and histopathology for nonmelanoma skin cancer (NMSC),13,16 but most clinical benefits of PDD were focused on BCC.
JEON ET AL
TABLE 2. Comparison of Surgical Features in Primary Squamous Cell Carcinomas Between the Aminolevulinic Acid (ALA) Photodynamic Diagnosis (PDD) and Methyl-Aminolevulinic Acid (MAL-PDD) Groups Characteristic Surgical margin, mm, mean SD Surgical margin, mm, n (%) ≤5 6~10 11~15 ≥16 Surgical depth, n (%) Subcutaneous layer Muscle fascia or muscular layer Periosteum or bony layer Mohs stage, mean SD Mohs stage, n (%) 1 2 3 ≥4
ALA-PDD group, n = 19
MAL-PDD group, n = 19
8 (42.1) 7 (36.8) 3 (15.8) 1 (5.3)
8 7 2 2
10 (52.6) 6 (31.6) 3 (15.8) 1.32 0.67
(42.1) (36.8) (10.5) (10.6)
9 (47.4) 6 (31.6) 4 (21.0) 1.42 0.84
15 (79.0) 2 (10.5) 2 (10.5) 0 (0.0)
14 (73.6) 3 (15.8) 1 (5.3) 1 (5.3)
p-value .09 .09 >.99 >.99 .63 .55 .36 .91 >.99 .67 .67 – .70 .63 .55 >.99
SD, standard deviation.
TABLE 3. Comparison of Surgical Margin and Mohs Stage of Low-Risk Squamous Cell Carcinoma According to Photodynamic Diagnosis (PDD) Application Non-PDD group Margin and stage
Mean standard deviation (n)
Low-risk location (site other than auricle, lip) Surgical margin, mm 10.84 6.14 Mohs stage 1.88 0.93 ≤20 mm in diameter Surgical margin, mm 11.86 8.80 Mohs stage 1.93 0.92 ≤4 deep Surgical margin, mm 12.44 8.25 Mohs stage 1.81 0.98 No ulcer Surgical margin, mm 13.62 8.84 Mohs stage 2.15 0.90 Low-risk histologic patterns (well-differentiated) Surgical margin, mm 12.44 7.81 Mohs stage 2.00 0.91
7.97 4.72 (31) 1.42 0.81 (31)
7.00 4.46 (28) 1.25 0.75 (28)
7.67 4.67 (21) 1.33 0.73 (21)
7.75 5.35 (20) 1.40 0.75 (20)
8.07 4.91 (28) 1.36 0.78 (28)
The relative risk of incomplete cSCC excision in MMS is significantly higher than that with BCC.3,17 The more irregular infiltration of SCC than BCC, in particular, by subclinical spread through small finger-like outgrowths from the tumor base, may cause this higher risk. Furthermore, this subclinical spread may occur more frequently in high- than in low-risk SCC.18 For this reason, a complementary
noninvasive technique such as PDD for the detection of SCC tumor margins would be helpful during the initial MMS step. In the current study, PDD using topical MAL or ALA increased the surgical efficacy of MMS by delineating the primary cSCC margins. The PDD group required fewer stages of Mohs excision and
PHOTODYNAMIC DIAGNOSIS IN SCC
TABLE 4. Comparisons of Surgical Margin and Mohs Stage of High-Risk Squamous Cell Carcinomas Depending on PDD Application (PDD Group vs Non-PDD Group) Non-PDD group Margin or stage
Mean standard deviation (n)
High-risk location (auricle, lip) Surgical margin, mm 13.75 11.15 (4) Mohs stage 1.50 0.58 (4) > 20 mm in diameter Surgical margin, mm 10.67 4.56 (15) Mohs stage 1.73 0.88 (15) > 4 mm deep Surgical margin, mm 9.77 4.44 (13) Mohs stage 1.85 0.80 (13) Ulcer Surgical margin, mm 9.31 3.98 (16) Mohs stage 1.56 0.81 (16) High-risk histologic patterns (moderately or poorly differentiated) Surgical margin, mm 9.27 4.50 (11) Mohs stage 1.55 0.82 (11)
p-value 8.29 6.65 (7) 1.14 0.38 (7)
10.90 5.63 (10) 1.70 0.67 (10)
8.47 5.54 (17) 1.41 0.80 (17)
8.33 4.77 (18) 1.33 0.78 (18)
7.90 5.61 (10) 1.40 0.70 (10)
PDD, photodynamic diagnosis.
smaller surgical margins for complete tumor removal than the non-PDD group (p < .05). Moreover, in the PDD group, 29 cases (76.3%) required one Mohs stage for complete tumor excision, whereas only 13 cases in the non-PDD group (44.8%) required one stage for complete tumor excision (p = .008). In the PDD group, 16 cases (42.1%) had a surgical margin of