Adjuvant Radiotherapy for Dissected Parotid Lymph Nodes Containing Merkel Cell Carcinoma from an Unknown Primary Skin Site: 2 Case Reports

Background

Merkel cell carcinoma (MCC) is an unusual primary skin tumor of the neuroendocrine type, with approximately 2500 cases per year in the United States [1]. MCC is one of the most malignant and aggressive skin cancers and has a poor prognosis and low survival rate, with high locoregional recurrence (LRR) and high distant metastatic (DM) rates of up to 25% in most series [2].MCC has a higher mortality rate than melanoma and generally occurs in people over 65 years of age. UV exposure and Merkel polyomavirus infection are major risk factors for MCC, and the most common locations of MCC are the head and neck (40%), extremities (33%), and trunk (23%). Immunosuppressed patients are more likely to develop MCC than age-matched controls[3]. Surgery has been the most common treatment for primary MCC tumors, and postoperative adjuvant treatments can lower locoregional recurrence rates and improve survival compared with surgery alone [4]. However, most studies on MCC treatments and outcomes were performed in Western countries, and there have been limited experiences in Asia. Therefore, we present 2 cases of metastatic MCC in the parotid lymph nodes without a detectable primary skin tumor treated by surgical excision followed by adjuvant radiotherapy, including 1case with a long-term follow-up of 10 years. Although some authors have questioned whether MCC cells only in lymph nodes without evidence of any primary cutaneous site can directly originate from lymphoid tissues, some laboratory findings have shown otherwise [5,6].We also review the current literature about treatments for MCC. We present the following case in accordance with the CARE reporting checklist.

Case Reports

CASE 1:

A 75-year-old male patient visited our ENT clinic on 2012/05/14 with a right preauricular mass that he noted 2 weeks prior. The tumor mass showed progressive slow growth associated with numbness of the right facial area. The patient was previously diagnosed with prostate cancer and diverticulitis, and he received definitive radiotherapy for the prostate tumor in our hospital in 2008.Physical examination of the patient revealed a fixed, round tumor mass with a 3 cm diameter and well-demarcated border located at the right preauricular area. A CT scan on 2012/6/14 showed a 2.7×1.5×1.7 cm well-encapsulated enhanced tumor at the tail of right parotid gland anterior to the right sternocleidomastoid muscle. The differential diagnosis included an exophytic mixed tumor (pleomorphic adenoma) of the right parotid tail or infection-related neck lymph-adenopathy. An FDG PET-CT scan on 2012/7/25 acquired with a Philips Gemini GXL-16 PET-CT system revealed multiple focal areas of increased FDG uptake involving the right parotid gland (SUVmax early: 2.5, delay: 3.0), right subclavian lymph node (LN) (SUVmax early: 2.4, delay: 3.8), right breast (SUVmax early: 3.0, delay: 2.0), and ascending colon (SUVmax early: 3.4, delay: 7.8). A tumor near the right ear and right parotid gland and right subclavian lymph node metastases were considered. The FDG-avid lesions in the right breast and ascending colon were due to the patient’s previous history of prostate cancer androgen deprivation therapy and sigmoid colon diverticulitis.

After all the exams were completed, the patient underwent right superficial parotidectomy with tumor removal on 2012/07/09. A firm, brown-gray cystic mass (2.8×2 cm)was found at the tail of the right parotid gland; the right greater auricular nerve traversing above the tumor was dissected and preserved.

Under microscopy, the pathology analysis revealed lymph node tissue with a small amount of salivary gland tissue in the pericapsular area; diffuse infiltration of small, plump, round, or irregular-nuclei in undifferentiated tumor cells, with scanty cytoplasm (high N/C) and extracapsular infiltration. Immunohistochemical staining was positive for CK20 and synaptophysin, and negative for TTF-1, proving metastatic MCC in 3 lymph nodes. Based on the above findings, the final diagnosis was metastatic MCC, and pathological staging performed according to the 8^th edition of the AJCC staging system showed pTXN1aM0, stage IIIA.

A referral was made to Dermatology for evaluation of 2 suspicious skin lesions. However, skin biopsies of 2 separate skin lesions on the right facial area revealed benign seborrheic keratosis. A whole-body skin assessment did not reveal any other skin nodules or lesions.

The patient received postoperative adjuvant radiotherapy (61.2 Gy in 34fx) to the tumor bed and right neck lymph nodes in levels IB, II, III, IV, VA, and VB and in the right subclavicular region from 2012/07/30 to 2012/09/18.The patient refused adjuvant chemotherapy. The dose to the remaining right parotid gland was within the constraints for a maximum dose, mean dose, and D50% of 45.72 Gy, 26.72 Gy, and 22.45 Gy. During long-term follow-up (10 years), the patient had mild xerostomia with no disease recurrence and underwent regular follow-up examinations in our ENT and radiotherapy oncology clinic.

CASE 2:

A 73-year-old female patient came to our ENT clinic on 2021/01/21 with an incidentally found left neck mass, and physical examination revealed a 2.5cm, movable, elastic, nontender mass over the left preauricular region. Fiberscopy showed a smooth nasopharynx and intact oropharynx and hypopharynx. A head and neck CT scan acquired on 2021/01/22 showed preserved parotid and submandibular glands, but 1suspicious mass lesion, measuring18×16×13mm, in the left parotid tail (Figure 1). Fine-needle aspiration was performed, and the report revealed necrotic exudate tissue mixed with diffuse leukocytes and a few atypical lymphocytes, but no malignant cells.

The patient underwent left parotid tumor excision on 2021/02/17. An elastic, brown-gray, lobulated parotid tail tumor measuring approximately 2.5×2 cm was found adhering to the sternocleidomastoid muscle posteriorly and superiorly, with the marginal mandibular nerve passing superiorly and retroauricular nerve posteriorly, confirmed by intraoperative nerve monitoring. The tumor was excised with the parotid gland with retraction of the facial nerve to reach the tumor from underneath.

Under microscopy, the pathological analysis showed lymph node tissue with a small amount of salivary gland tissue in the pericapsular area. The tumor cells were closely spaced in sheet-like and trabecular pattern; frequent mitosis was observed, with no extracapsular infiltration. Immunohistochemical staining showed CK20+, synaptophysin+, and a high proliferative index(Ki67: 70%) (Figure 2). According to the above histo-pathological findings, the final diagnosis was metastatic lymph node with MCC without a primary skin tumor, and pathological staging performed according to the 8th edition of the AJCC staging system showed pTXN1aM0, stage IIIA.

We also acquired a postoperative PET/CT scan on 2021/02/22 for further assessment, and the results revealed ill-defined FDG avidity in the nasopharynx (SUVmax 4.38/6.80; score 1) and bilateral tonsils (SUVmax 6.03/8.35; score 1), and an ill-defined FDG-avid lesion in the left parotid region (SUVmax 4.00/5.04; score 1), which was compatible with the malignant left parotid node obtained postoperatively, with inflammation, and suggestive of no apparent FDG-avid nodal or distant metastasis (cN0M0).

This patient also received postoperative adjuvant radiotherapy. She visited our radiation oncology clinic on 2021/03/17. We administered adjuvant radiotherapy using 6MV X-ray (RapidArc-Varian, Palo Alto, CA) to the CTV, including the left parotid tumor bed and left neck lymph nodes at level IB, II, III, IV, VA, and VB with 56 Gy in 28fx, during 2021/03/23-2021/04/30 (Figure 3). We checked the treatment site with daily cone-beam CT scans. The dose to the remaining left parotid gland was within constraints for a maximum dose, mean dose, and D50% of 60.11 Gy, 18.26 Gy, and 11.09 Gy, respectively. The patient tolerated the treatment well and experienced grade II xerostomia after the whole course of radiotherapy. The patient was alive and disease free for 14 months after completing radiotherapy and was put on oral Salagen to treat her dry mouth.

Discussion

DIAGNOSIS:

MCC was first described in 1972, when Toker presented the first 5 cases under the name “trabecular carcinoma of the skin”. Merkel cells were also reported as “touch cells” based on their presumed association with nerve fibers in the skin. MCC is more common in Western countries than in Asia. The clinical presentation of MCC is usually nonspecific and varied, but the most typical expression is a red or purple nodule on sun-exposed areas of the head or neck region in elderly people. Sun exposure and Merkel polyomavirus infection have both been implicated in the development of MCC [7].

The diagnosis of MCC usually requires biopsy with immunohistochemical staining due to its nonspecific clinical presentation and histopathologic appearance. MCC tumors typically contain small, round, blue cells with little cytoplasm, frequent mitoses, and dense core granules in the cytoplasm. Cytokeratin20 (CK20), a low-molecular-weight cytokeratin, commonly expressed in colorectal, pancreatic, gastric, and gallbladder carcinoma, is also a highly sensitive marker for MCC. Thyroid transcription factor (TTF)-1 is usually expressed in the epithelial cells of the lung and thyroid but is consistently absent in MCC. Therefore, positive CK20 and negative TTF-1 expression would establish a diagnosis of MCC [8].Other immunohisto-chemistry neuroendocrine markers, including synaptophysin, chromogranin, CD56, and neuron-specific enolase, might also be positive in MCC, but these markers have low specificity [9]. The pathology reports of these 2 patients showed positive CK20(+++, >100% in dot-like pattern), positive synaptophysin(+++, 100% in membranous pattern), and negative TTF-1, leading to the diagnosis of MCC.

IMAGING:

Because nodal metastasis portends a poor prognosis, it is crucial to accurately assess nodal status in MCC. Thorough physical and clinical examination of the patient, and other complementary tools such as computed tomography (CT), magnetic resonance imaging(MRI), positron emission tomography (PET), and sentinel lymph node biopsy are all recommended for evaluating nodal status. Only a few studies have evaluated CT in the detection of MCC tumors, but similar reports have shown that this modality has a low sensitivity (30–47%) and high specificity (87–97%) in detecting nodal disease[10,11]. In contrast, FDG PET/CT has high sensitivity (90%) and specificity(98%), suggesting that FDG PET may be more useful than CT in evaluating nodal and distant metastasis in patients with MCC. This is similar to observations for other metastatic skin cancers, although FDG metabolism may be affected by previous radio-therapy that could subsequently elicit antitumor immune responses[12,13]. In our first case, the PET scan showed a right subclavian lymph node metastasis that was not seen on the CT scan, thus proving the usefulness of PET scans in the work-up for MCC metastases.

NODAL STAGING:

The AJCC divides regional LN staging into clinical(cN) and pathologic (pN) stages. Pathologic nodal evaluation improves prognostic predictions, and sentinel LN (SLN) status can help to guide management and improve regional control. The incidence of occult lymph node metastasis is approximately 30% in clinically node-negative MCC patients [14]. Therefore, sentinel LN biopsy is recommended for all patients with clinically node-negative MCC who are suitable for surgery. Both of our patients underwent parotidectomy with tumor excision, and metastatic MCC in the lymph nodes was found postoperatively in both cases. This also means that potential lymphatic metastasis always needs to be suspected in MCC, as this type of metastasis is more frequent in MCC than in other aggressive skin cancers, such as squamous cancer [15].

TREATMENT:

The treatment of MCC typically depends on the primary tumor and extent of the disease, especially the presence of lymphatic or distant metastasis. For MCC patients, surgery with wide local excision or Mohs micrographic surgery (MMS) is the mainstay of treatment for the primary lesion. Margin status has an impact on survival, and 1–2 cm margins are recommended [16]. For positive margins, re-excision with a tumor-free margin should be considered. If other adverse risk factors exist, including larger tumor size (>1 cm), immuno-compromised status, and lymphovascular invasion, adjuvant RT should be considered. Because of the complex and closely confined neurovascular anatomy of the head and neck, primary tumors located in the head and neck are often complicated with high-risk features.

Many studies have shown that postoperative adjuvant RT to the tumor bed is associated with lower locoregional recurrence rates and/or improved survival compared with surgery alone. One study showed that 42 stage I and II MCC patients treated with surgery and adjuvant RT to the primary tumor bed had a 2-year local recurrence-free survival rate of 89% compared with36% for patients who did not receive adjuvant RT (p<0.001) [17]. The cumulative 2-year regional recurrence-free survival rate for patients who received adjuvant regional RT was 84% compared with 43% for patients who did not receive adjuvant RT (p<0.001).

In one study,112 MCC patients received surgery with postoperative adjuvant radiotherapy to the primary tumor bed and regional lymph node area. The 2-year locoregional control rate was 75%, and the 2-year and 5-year overall survival rates were 72% and 53%, respectively. The study also showed an in-field relapse rate of 3% for primary disease [18]. Patients who received more than 50 Gy had a significant lower relapse rate (hazard ratio [HR]=0.22; 95% confidence interval [CI], 0.06–0.86). However, patients who did not receive elective postoperative adjuvant RT to the lymph nodes had a much higher rate of nodal recurrence than those who did (HR=6.03; 95% CI, 1.34–27.10). This study found suggested doses greater than 50 Gy for sub-clinical disease and more than 55 Gy for gross tumor. Primary tumors and the draining nodal basin should be treated in all MCC patients. Our first patient was treated according to the NCCN clinical practice guidelines for MCC. In total, 60 Gy was delivered to the regional lymph nodes in the right neck because the patient had clinically evident lymphadenopathy but did not undergo lymph node dissection, and 60 Gy was delivered to the preauricular primary site since there was extracapsular infiltration. The second patient was treated with 56 Gy to the tumor bed and regional node area to prevent locoregional recurrence despite no apparent node metastasis and no extracapsular spread.

For patients who are poor surgical candidates or refuse surgery, definitive RT might be an alternative treatment that can provide good outcomes. MCC is a highly radiosensitive tumor, which has been demonstrated in in vitro studies [19]. A retrospective study that included 57 patients who received definitive radiotherapy alone to the primary and/or nodal sites reported that the tumor control rates for the gross tumor were 88% at 1 year and 82% at 2 years, and the 5-year local relapse-free survival rate was 90%. The 5-year relapse-free survival, cancer-specific survival, and overall survival rates were 57%, 68%, and 39%, respectively [20].

In a series of microscopically lymph node-positive MCC cases, the regional control rate was 100% irrespective of the treatment consisted of definitive lymph node irradiation alone(n=19) or complete lymphadenectomy+adjuvant radiotherapy (n=7), with a median follow-up of 18 months[21]. Patients with clinically positive lymph nodes by gross observation had 2-year regional recurrence-free survival rates of 78% and 73% in the definitive lymph node irradiation alone(n=9) and complete lymphadenectomy+adjuvant radiotherapy (n=15) groups, respectively (p= 0.8, median follow-up of 16 months). The study showed that lymph node irradiation alone to the microscopic and macroscopic positive regional lymph nodes conferred an excellent regional control rate that was comparable to that of lymphadenectomy. Another study proved that lower dose ranges (30/40 Gy in 2 Gy/day/fractions) were sufficient to stably control extensive skin involvement of large body areas when symptom palliation was the primary goal of radiotherapy [22].

In our 2 cases, we used photon-based radiotherapy due to its great versatility in simultaneously achieving optimal dose coverage of both the subcutaneous primary tumor bed and deeper draining lymph node basins. However, electron beam radiotherapy is also effective for treating skin cancers [23] and seems to be a reasonable option for head and neck MCC treatment to limit the risk of severe oropharyngeal mucositis in frail patients [24].

Most of the studies in which some subsets of MCC patients received surgery and postoperative chemotherapy, often in conjunction with adjuvant RT, showed that the use of chemo-therapy did not lower the risk of recurrence or distant metastasis, and did not improve survival [25].Platinum-based chemotherapy generally only led to transient responses [26] in MCC, and targeted therapies have not been successful [27]. In recent years, great breakthroughs have been made in programmed death ligand 1 (PD-L1) and programmed death receptor 1 (PD-1) immune checkpoint inhibitors for MCC. For the treatment of advanced MCC, avelumab was associated with an objective response rate (ORR) of 39.7% (95% CI 30.7–49.2) as first-line therapy and 33% (23.3–43.8%) in patients with chemotherapy-refractory metastatic MCC, with durable responses [28,29]. Pembrolizumab was associated with an ORR of 58% (95% CI 43.2–71.8) in a phase 2 study of 50 patients with advanced MCC[30]. These 2 patients with MCC lymph node metastasis underwent tumor resection and received adjuvant radiation therapy without further chemotherapy or immunotherapy, and there has still been no evidence of recurrence. Systemic therapy with chemotherapy or immunotherapy may be considered for the treatment of patients with disseminated MCC.

Conclusions

The management of MCC requires input from a multidisciplinary team of experts to achieve optimal patient outcomes. In our experience, metastatic MCC of the parotid lymph nodes without a detectable primary skin tumor is very rare. Adjuvant radiotherapy to the tumor bed and regional nodal basin might be beneficial for preventing disease recurrence despite the absence of systemic medical therapy. In the future, as with other cancer treatments, we need to accumulate more case experiences to more fully understand the direction of MCC treatment.