Feasibility of Sentinel Lymph Node Biopsy by Intralesional Injection in Marjolin Ulcer

In the first century CE, Celsus noted that cancer develops in burn scars and chronic ulcerating wounds.^1,2 French surgeon Jean-Nicolas Marjolin called such lesions ulcers canceroides in 1828, and the lesion now bears his name.³ Marjolin offered that “all types of ulcer can become canceroidal being irritated. The results obtained by treatment usually clarified the diagnosis.”³ 

In 1930, Treves and Pack⁴ concluded that approximately 2% of burn scars will have some malignant changes.^1,2 The reported incidence rates for the origin of the lesion were 38% in lower extremities, 22% in upper extremities, 30% about the head, and 10% in the trunk. The authors of a 2008 article reported Marjolin ulcer to be associated with mortality and morbidity rates higher than those for the typical squamous cell carcinoma (SCC).⁵ Marjolin ulcer occurs in unexcised full-thickness burns an average of 30 years after injury.⁶

Approximately 150 years after Marjolin reported his findings, Cabanas⁷ proposed the sentinel lymph node (SLN) concept and hypothesized that the most common site of metastatic lesion of penile SCC would be an SLN and, if such a node were histologically positive for metastasis, it would necessitate traditional lymphadenectomy.Sentinel lymph node biopsy (SLNB) in the evaluation of skin malignancies is a minimally invasive means of staging and identifying regional nodal metastases. Taking into account the ideas of both Marjolin and Cabanas, some authors recommend using SLNB in the assessment and staging of Marjolin ulcer.²

The SLN idea proved that lymphatic mapping intraoperatively and then selective lymphadenectomy would improve the accuracy of staging and change the management of SCC and malignant melanoma.⁸ This method has led to the use of SLNB and selective lymphadenectomy as a good substitute to elective lymph node dissection in evaluating patients with possible microscopic metastasis.²

Considerable variability exists among different institutes in the evaluation of and approach to management of regional metastases to lymph nodes, however.² The optimal approach to regional lymphatic basin in the patient with Marjolin ulcer is unclear. Some experiences with SLNB in the management of this malignancy transcend previously described advantages of this approach.²

The SLNB procedure is performed using technetium Tc 99m nanocolloid as a radioactive tracer and patent blue dye for lymphatic mapping. Each material is injected into the tissue surrounding the tumor or the biopsy cavity or, in the case of breast cancer, into the periareolar area. The SLN is located using a triple technique consisting of lymphoscintigraphy, intraoperative gamma probe, and detection of the blue lymphatic vessels and node intraoperatively.^9,10

In breast cancer, the complication rate of formal axillary lymph node dissection is approximately 30%.⁹ The technique of SLN sampling helped reduce this rate to 5%.⁹ A patient with Marjolin ulcer typically has a history of a burn injury, often followed by skin grafting and resulting joint contracture (Figure 1). These problems would intensify the morbidity of complete lymph node dissection, and their presence supports use of a selective approach to avoid the complications of a formal regional lymphadenectomy.²

Regional lymph node dissection should be considered in patients with positive nodes after first ruling out distant metastatic disease. Management of the clinically negative lymphatic basin is more controversial. In patients with traditional SCC, SLNB has been demonstrated to be accurate and is recommended for consideration in patients at higher than average risk of nodal metastases.¹ Accurate assessment of nodal metastases affects adjuvant treatment plans, so it would be reasonable to consider SLNB in the patient with Marjolin ulcer. Given the accuracy of SLNB, there is no clear role for elective lymph node dissection.¹¹

No study has been done specifically on Marjolin ulcer, but a case report of 6 patients showed no clinical lymphadenopathy; however, in that study 80% of SLNBs revealed evidence of nodal involvement.² The prognostic value of this finding is unclear, although confirmation of positive lymph nodes might result in earlier diagnosis of Marjolin ulcer. It is possible that early and precise staging might allow improvement in the treatment approach to these subtle regional nodal metastases.²

In the current study, the authors assessed the validity of SLNB by means of tumoral injection of the radiotracer and patent blue dye in Marjolin ulcer. 

This study included 9 patients with Marjolin ulcer who were candidates for surgical removal of their tumor. The study was conducted from April 2008 through December 2010. The underlying condition for all patients was burn wounds. Consecutive patients were included in the study. Injection of the radiotracer and lymphoscintigraphy were performed at the nuclear medicine department at Mashhad University of Medical Sciences 2 hours to 4 before surgery. Patients received 2 mCi technetium Tc 99m antimony sulfide colloid in 2 divided doses of 0.1 mL each in an intratumoral fashion. All patients had extensive burn scars in the peritumoral area, and material was injected into the tumor. Lymphoscintigraphy of the regional lymphatic basin was performed 30 minutes after injection in anterior-posterior projections (ie, 128 × 128 matrix size, Tc 99m photopeak, and low-energy high-resolution collimator using a dual-head variable-angle gamma camera [E.Cam; Siemens]). Next, patients were moved to the operating room. Immediately after anesthesia induction, 1 mL of patent blue dye in 2 divided doses was injected in the same fashion as the radiotracer. Lymphatic mapping and SLNB were performed using a gamma probe (Europrobe; Eurorad). Any hot and/or blue lymph node was harvested as an SLN. Primary lesions were then excised and split-thickness skin grafting performed. Pathologic examination of the harvested sentinel nodes was done using frozen section, step sectioning, and hematoxylin-eosin staining. Patients underwent regional lymph node dissection. 

Written informed consent was obtained from all participants before the study commenced, and the study was approved by the local ethical committee of Mashhad University of Medical Sciences.

Nine patients with Marjolin ulcer (1 ulcer per patient) were included in this study (7 male, 2 female). Patient age range was 24 to 75 years (mean, 52 years ± 17.02 standard deviation). Time from burn wounding to the development of Marjolin ulcer was 20 to 70 years. The average wound size was 7.2 cm².

Two lesions occurred in the upper extremity, and 7 occurred in the lower extremity. In all 9 patients, the pathology of the primary ulcer was well-differentiated SCC.

Lymphatic mapping was successful in 5 patients. The details of sentinel node mapping results of the patients in the current study are provided in the Table. As shown in the Table, SLN was identified on scintigram in only 4 patients (Figure 2).

All samples were initially assessed by frozen section and then by permanent pathologic examinations, all of which were negative. Regional lymphatic dissection was performed in 8 of 9 patients as per the decision of the surgeon in charge. These samples also were negative for malignancy.

Cutaneous SCCs (CSCCs) arising from sites of chronic skin inflammation are among the most aggressive skin cancers,¹² so it is important to consider Marjolin ulcer in any case of chronically inflamed skin or a nonhealing wound. It is important to maintain a low threshold for biopsy and further evaluation.¹¹ Squamous cell carcinoma in patients with a history of radiotherapy or chronic skin inflammation has been shown to have a high rate of regional nodal involvement, distant metastasis, and high overall mortality.¹³ 

Definitive surgical management includes wide local excision with 2-cm to 4-cm margins, and SLNB should be considered. Given the aggressive nature of Marjolin ulcer, sufficient local control and systemic staging is critical to optimize outcomes.¹¹

The 5-year and 10-year survival rates of patients with CSCC with regional nodal involvement are 30% and 15%, respectively.^14,15 Early detection of regional nodal involvement in patients with clinically node-negative SCC is of prognostic significance and can guide the managing physician to pursue more aggressive therapies, such as regional nodal dissection and radiotherapy, with the potential for an improved rate of survival.¹⁶

To the authors’ knowledge there are limited published data on lymphatic mapping in Marjolin ulcer. The few articles identified show that SLNB has the potential to evaluate the regional lymph nodes in patients with Marjolin ulcer,² which is borne out in the current study. In the SLN technique, a radiotracer such as technetium-labeled radiocolloid or a blue dye (eg, isosulfan blue) injected around a tumor primary site drains into the SLN. In tumors such as melanoma, if the SLN is free of tumor, the possibility of micrometastasis in the other lymphatic basin is less than 1% to 2%⁹; thus, SLNs may be the primary hosts of metastatic cells in some patients. If this is the case, then patients with Marjolin ulcer with noninvolved SLNBs are at very low risk of cancer recurrence in the defined lymphatic basin.² 

In one case report, SCC occurred a median of 29.5 years after burn wounding (range, 20–32 years).² In the current study, the range of time from burn to SCC was considerably larger (20–70 years). 

Most studies on Marjolin ulcer, such as this study, include few cases. The prevalence of Marjolin ulcer is so low that statistical analysis of predicting factors leading to metastases (regional or distant) is not possible. There is also doubt concerning whether or not a positive SLNB predicts cancer-related complications and morbidity.²

An article published in 2001 reported on 56 patients with Marjolin ulcer treated over a 16-year period, for an average of 3.5 cases per year.¹⁷ In a study published in 2018 reporting on 153 SLNs in 88 patients, micrometastases were found in 3 patients with high-risk CSCC.¹⁸ The overall rate of micrometastases was 3.4% and was correlated with lesion diameter and depth. In patients with high-risk CSCC the rate of micrometastases was 6.5%.

In another study, sentinel node mapping was performed for 31 patients with CSCC at high risk for metastasis, with 54.8% of these patients exhibiting positive sentinel nodes on radiologic examination.¹⁹ These nodes were negative for metastasis, but at later follow-up lymph node metastases were found to have developed in 2 patients (6.5%).

The incidence of Marjolin ulcer in the current study population was similar to that reported in the aforementioned studies. 

None of the cases presented in the literature reported the extent of any patient’s underlying condition or disease from which the malignancy arose. 

The negative pathology in this study fails to support or deny these results because of the rarity of the Marjolin ulcer case.

In approximately 45% of patients with a Marjolin ulcer, the primary lesion involves the trunk or the head and neck; however, lymphatic drainage is unpredictable in most such cases.² In the current study, the authors used data on extremity lesions.  

Patients with Marjolin ulcer often undergo excision of the burned area and partial-thickness or, in some cases, full-thickness skin grafting procedures, which can alter cutaneous lymphatic drainage. Most cases of Marjolin ulcer occur in patients with vast burn areas involving greater than 40% of body surface area. In some studies, the size of the primary burn lesion resulted in failure of mapping.² In the case study by Eastman et al,² preoperative lymphoscintigraphy allowed for intraoperative diagnosis of the lymph nodes as well as the number of SLNs. 

High-spatial-resolution T2-weighted magnetic resonance imaging in the axillary region showed acceptable specificity regarding validation between each node and every patient.²⁰ It guarantees good interobserver agreement. The negative predictive value of this test, however, is not yet sufficient for it to replace SLNB in ruling out axillary lymph node metastasis.²⁰ Currently, some trials question SLNB in breast cancer.⁸ 

In patients with ovarian cancer, injecting a tracer into the ovarian ligament has resulted in the detection of sentinel nodes. One study compared intraoperative sentinel node detection with detection on postoperative single-photon emission computed tomography/computed tomography (SPECT/CT).²¹ Differences between the results of gamma probe and SPECT/CT can be caused by missed sentinel nodes during surgery. In many cases it seems that pelvic hotspots are related to tracer remnants at injection sites. Considering the location where sentinel nodes were obtained in the study by Speth et al,²¹ residual hotspots may be due to remaining lymphatic flow after resection.

In contrast to patients with basal cell carcinoma, patients with CSCC should undergo lymphoscintigraphy, because nodal metastases are possible.¹⁹ Proper diagnosis and management will reduce morbidity and mortality rates.

In a study published in 2015, lymphatic mapping and SLNB using peritumoral radiotracer injection did not prove to be feasible in Marjolin ulcers located in areas of extensive burn scarring.¹³ 

In this study a novel means of identifying sentinel node in patients with Marjolin ulcer at the site of preexisting burn wounds was used. Because Marjolin ulcer is rare and so few patients were studied, results data are limited. 

The results of this study suggest that damaged tissue secondary to a chronic wound (eg, on a burn scar) results in less developed lymphatic tissue surrounding a Marjolin ulcer, which interferes with the movement of radiotracer in the lymphatics and thus leads to a low detection rate. All 9 patients included in this study had extensive burn scars, with no healthy skin near the malignant lesions; consequently, the radiotracer and blue dye were injected into the scar tissue of the tumor.

The extent of underlying disease can be associated with failure of sentinel node detection and lymphatic mapping. With a less extensive condition, injection of the dye or radiotracer may be done in less involved or noninvolved areas of the skin with the chance of effective sentinel node detection. As mentioned previously, the patients in the current study had experienced extensive burn injury and all injections were made in the tumor itself. It seems that for Marjolin ulcer on an extensive burn scar, it is better not to rely on peritumoral lymphatic flow for lymphatic mapping. However, more extensive prospective randomized trials are necessary to further study this topic.

Authors: Ezatollah Rezaei, MD^1,2; Mansoureh Sargazi Moghadam, MD¹; Ramin Sadeghi, MD³; and Amin Jahani, MD, RIP^1,2

Affiliations: ¹Endoscopic and Minimally Invasive Surgery Research Center, Mashhad University of Medical Sciences, Mashhad, Iran; ²Ghaem Hospital, Mashhad University of Medical Sciences, Mashhad, Iran; ³Nuclear Medicine Research Center, Mashhad University of Medical Sciences, Mashhad, Iran 

Correspondence: Mansoureh Sargazi Moghadam, MD, Endoscopic and Minimally Invasive Surgery Research Center, Mashhad University of Medical Sciences, Mashhad, Khorasan Razavi, Iran, 9188746596; msargazim@gmail.com 

Disclosure: The authors disclose no financial or other conflicts of interest.