SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusion
  7. References

Background

Merkel cell carcinoma (MCC) is an aggressive cutaneous malignancy with a high rate of nodal metastasis. The American Joint Committee on Cancer staging system subclassifies nodal disease into microscopic and macroscopic groups based on prognosis.

Objective

To compare the outcome of patients with microscopic and macroscopic nodal metastases.

Materials and Methods

Patients were identified from a database of 180 patients with MCC who presented to Westmead Hospital, Sydney, Australia, from 1980 to 2013. Disease-free survival (DFS), overall survival (OS), and follow-up were calculated using Kaplan-Meier curves and compared using the log-rank (Mantel-Cox) test.

Results

Forty-one patients were diagnosed with node-positive MCC; 11 patients had microscopic nodal metastases, with five (45%) relapsing, and 30 had macroscopic disease, with 17 (57%) relapsing. There was no significant difference in DFS (= .93) or OS (= .63) between the two groups.

Conclusion

The nonsignificant difference in DFS and OS suggest that even microscopic nodal metastases can predict a poor outcome. Because more than half of patients subsequently relapse, often at a distant site, there is a need to develop an effective systemic treatment.

Merkel cell carcinoma (MCC) is an aggressive cutaneous malignancy first described by Toker in 1972.[1] Although it is rare, with an age-adjusted incidence in the U.S. Caucasian population of 0.44 in 100,000 in 2001, its incidence is increasing.[2] MCC arises slightly more frequently in men and mostly in older (>70) Caucasians.[3] Factors postulated to contribute to the etiology of MCC include long-term sun exposure and immunosuppression, with a markedly higher incidence (10–30 times) in patients with hematologic malignancy and human immunodeficiency virus and in solid organ transplant recipients.[4-6] The discovery of the Merkel cell polyomavirus (MCV) has been implicated in the pathogenesis and possibly affects the prognosis of patients.[7, 8]

Merkel cell carcinoma usually presents as a rapidly progressing, pink–purple, cutaneous, nonulcerative nodule, typically on sun-exposed areas of the skin such as the head, neck, and extremities. Most patients present with a primary lesion without clinically detectable nodal metastases, with approximately 30% having palpable nodal metastases and only approximately 5% with distant metastatic disease.[9] Sentinel lymph node biopsy (SLNB) is increasingly used to investigate patients who present with clinically node-negative disease, with reported rates of a positive SLNB of 30% to 50%.[10-12]

The most-recent version of the American Joint Committee on Cancer (AJCC) staging system for MCC, based on the results of an analysis of more than 5,000 patients, reported a significant difference in survival between patients with microscopic (or occult) and macroscopic nodal metastases.[8]

At our institution, we are increasingly using SLNB to stage patients who present with Stage I or II (clinically node negative) MCC. With the finding that many of these patients have microscopic nodal metastases (Stage IIIa), we were interested to determine whether their outcomes differed from those of patients with clinical nodal metastases (Stage IIIb). The aim of this study was therefore to compare the outcome of patients with microscopic and macroscopic nodal disease treated at our institution.

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusion
  7. References

The Westmead Hospital Ethics Committee approved this study. Patients who presented with macroscopic nodal metastases or were diagnosed with micrometastatic metastases after undergoing a SLNB were identified from a prospectively maintained database of 180 patients with Stage I to III MCC who presented to Westmead Hospital, Sydney, Australia, between 1980 and 2013. Data on patient demographic characteristics, tumor and treatment details, and patient follow-up were extracted from a computer database and patient medical records. Patients with metastatic nodal disease of unknown primary site were excluded from the study.

Treatment

All patients underwent treatment of the primary site and the involved nodal basin with surgery, radiotherapy (RT), or both. Our unit has increasingly used SLNB in MCC since 1998, and it is currently offered to all patients with clinically node-negative disease who are fit for surgery. Informed consent was obtained from patients before the procedure. Preoperative lymphoscintigraphy was performed in which Technetium99 sulphide colloid was injected intradermally in four quadrants around the primary tumor or biopsy scar. Dynamic planar imaging was then performed using a standard gamma camera, followed by single-photon emission computed tomography. All patients then proceeded to surgery under general anaesthesia.

Intraoperatively, all patients had an intradermal injection of 2 mL in total of patent blue dye (Laboratoire Guerbet, Aulnay-Sous Bois, France) in four quadrants around the primary tumor or biopsy scar. The sentinel node(s) was (were) located using a gamma probe and the presence of the patent blue dye, and then resected and sent for histopathologic examination. Each sentinel lymph node was serially sectioned and then stained with hematoxylin and eosin and immunohistochemistry stained with CK20 and AE1/AE3.

Adjuvant RT was recommended at the primary site if the surgical margins were positive or close (<10 mm) with RT field margins of 3 to 4 cm (respecting organs at risk, e.g., eye), ideally to encompass in-transit lymphatics. All patients with microscopic nodal metastases and 80% of those with macroscopic nodal metastases were treated using RT with or without completion lymphadenectomy. Where possible, the draining nodal region was irradiated en bloc with the primary site and in-transit tissue. In patients with extremity lesions, this was often not possible, and separate RT fields were required. Doses of 50 to 55 Gy in 20 to 25 fractions were prescribed, often using an ipsilateral moderate (9–12 MeV) energy electron field (head and neck) or two field (AP/PA) megavoltage photon beams (axilla and groin).

After completion of treatment, patients were followed up every 2 to 3 months for 2 years and every 4 to 6 months thereafter.

Statistical Analysis

Statistical analysis was performed using Graphpad Prism 6 (San Diego, CA). Survival and follow-up were calculated using Kaplan-Meier analysis, and disease-free survival (DFS) and overall survival (OS) curves were compared using the log-rank (Mantel-Cox) test.

Results

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusion
  7. References

Patients and Tumor Details

Thirty patients presented with macroscopic nodal metastases with an identifiable primary lesion present, and 11 were diagnosed with micrometastatic nodal metastases after SLNB. Table 1 summarizes patient demographic, tumor, and treatment details.

Table 1. Patient and Tumor Details
CharacteristicMicroscopic Nodal Metastases, n = 11Macroscopic Nodal Metastases, n = 30
Age, median (range)72 (37–88)78.5 (49–96)
Sex
Male7 (64%)17 (57%)
Female4 (36%)13 (43%)
Immunosuppressed2 (18%)2 (7%)
Primary site
Head and neck3 (27%)16 (53%)
Limb6 (55%)10 (33%)
Trunk2 (18%)4 (13%)
Primary size, mm, median (range)12 (7–38)16.5 (8–100)
Number of nodes involved, median (range)1 (1–27)1 (1–5)

Median age at diagnosis and sex were similar between the two groups. A greater proportion of patients with microscopic metastases were immunosuppressed (18%) (chronic lymphocytic leukemia, cardiac transplant) than in the macroscopic group (7%) (renal transplant, non-Hodgkin's lymphoma). The most-common site of the primary tumor was the head and neck in patients with macroscopic metastases, whereas the limb was the most-common site in patients with microscopic metastases.

Median primary tumor size was slightly larger in the group with macroscopic nodal metastases, and median number of nodes involved was equivalent for each group.

Treatment

In each group, most were treated with a combination of surgery and RT. No patients with microscopic nodal disease were treated with surgery alone, and almost one-third of each group was treated with RT alone. In contrast, surgery was the sole modality in six of the patients with macroscopic nodal disease (Table 2).

Table 2. Nodal Treatment
TreatmentMicroscopic Nodal MetastasesMacroscopic Nodal Metastases
N (%)
Surgery alone0 (0)6 (20)
Radiotherapy alone8 (73)16 (53)
Surgery and radiotherapy3 (27)8 (27)
Total1130

Relapse

The relapse rate was greater in patients with macroscopic nodal MCC (57%) than in the microscopic nodal group (43%). No patients with microscopic nodal disease relapsed in the treated nodal basin, and the rates of local and distant relapse were similar between the two groups (Table 3).

Table 3. Relapse
Site of relapseMicroscopic Nodal Metastases, n = 11Macroscopic Nodal Metastases, n = 30
N (%)
None6 (55)13 (43)
Local1 (9)3 (10)
Regional0 (0)6 (20)
Distant4 (36)8 (27)

Survival

Median DFS was 19 months for patients with microscopic nodal MCC and 14 months for those with macroscopic nodal MCC (= .93) (Figure 1), and median OS was 34 months for patients with microscopic nodal metastases and 51 months for those with macroscopic nodal metastases (= .63) (Figure 2).

image

Figure 1. A comparison of disease-free survival between patients with microscopic nodal metastases and those with macroscopic nodal metastases showing no significant difference between the two groups.

Download figure to PowerPoint

image

Figure 2. A comparison of overall survival between patients with microscopic nodal metastases and those with macroscopic nodal metastases showing no significant difference between the two groups.

Download figure to PowerPoint

Median follow-up was longer for patients with macroscopic nodal MCC (48 vs 30 months). At the time of last follow-up, approximately half of each group was alive without disease, and a slightly greater proportion of the macroscopic nodal disease group had died from disease than of the microscopic nodal group (Table 4).

Table 4. Status at Time of Last Follow-Up
StatusMicroscopic Nodal Metastases, n = 11Macroscopic Nodal Metastases, n = 30
N (%)
Alive with no disease6 (55)14 (47)
Alive with disease1 (9)2 (7)
Dead of disease3 (27)12 (40)
Dead of other causes1 (9)2 (7)

Discussion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusion
  7. References

Patients diagnosed with MCC have a high propensity to develop regional metastases, many of which are subclinical (microscopic or occult). The finding of nodal metastases is considered a powerful independent predictor of outcome and often a portent of subsequent distant relapse and death. A large analysis of 5,823 patients with MCC found that patients with microscopic nodal metastases had a significantly greater likelihood of 5-year survival (42%) than those with macroscopic nodal metastases (26%).[8] These findings formed the basis for the most-recent edition of the AJCC staging system for MCC, in which patients who presented with nodal metastases were stratified into those with microscopic (stage IIIa) and macroscopic nodal metastases (stage IIIb). In concordance with this stratification, a study of 500 patients with MCC treated at the Memorial Sloan-Kettering Cancer Center (MSKCC) also demonstrated a significant difference in disease-specific death between patients with stage IIIa and IIIb MCC.[13]

We analyzed our data and, in contrast, were unable to demonstrate a significant difference in DFS or OS. In view of other published series, this was an unexpected result and may reflect factors such as the higher proportion of immunosuppressed patients with microscopic nodal metastases, a difference in follow-up time, and other factors such as tumor depth[14] and lymphovascular invasion,[15] which have been associated with survival in other studies and were not examined in this study. Primary tumor site also differed between the two groups of patients in our study, with the head and neck more common in patients with macroscopic nodal disease. Bajetta et al. reported that patients with a primary tumor located in the head and neck had a significantly higher recurrence rate,[16] although a larger review comparing head and neck with non-head and neck primary tumors found no difference in disease-specific survival.[17] Because our study had small patient numbers, it is unclear how much the differences between the groups contributed to the outcome.

Another limitation of our study was the difference in demographic characteristics and treatment between the two groups of patients. The treatment approach to MCC has evolved since 1980, with increasing evidence regarding benefit of RT.[18-20] In our study, six patients with macroscopic nodal disease were treated with surgery alone, whereas all of the patients in the microscopic group received RT as definitive or adjuvant therapy. This largely reflects the increasing use of RT more recently because five of the six patients who underwent surgery alone were treated before 2000. The primary tumor site influenced the approach to RT, with patients with a primary located in the head and neck having the nodal basin treated en bloc with the primary tumor where feasible. This may have affected patient outcome because a greater proportion of patients in the macroscopic group had a primary tumor located in the head and neck region than of those in the microscopic group.

Patients with metastatic nodal metastases with an unknown primary were excluded from our study because they have been shown to have a much better prognosis than those with a documented primary lesion,[21, 22] and thus their inclusion may have influenced survival of the group with macroscopic metastatic nodes. The reason for the better outcome in this group is unclear. Both of our groups consisted of mainly elderly patients (median age >70), and of those who died, the majority did so from MCC rather than another cause, suggesting that OS, at least in our cohort, was a more-reasonable outcome measure than DSS. Fewer than 10% of our patients died of unrelated causes, which contrasts with data from MSKCC, where they documented that one-quarter of MCC patients died of other causes.[13]

At our institution, SLN status influences subsequent management, with patients with evidence of metastatic nodal disease undergoing further treatment to the nodal basin.[10] The presumption is that a positive SLNB may be associated with further microscopic MCC in the remaining nonsentinel nodes, although data to support this assumption are limited. In the current study, all patients with microscopic nodal metastases underwent adjuvant RT to the nodal basin. Because none of these patients developed nodal relapse, this would suggest that RT is an effective method of achieving regional control in patients with microscopic metastatic metastases. In contrast, six (20%) patients with macroscopic nodal metastases developed nodal recurrence, with five having had RT alone or in combination with surgery. Similarly, in a study by Fang and colleagues that documented regional treatment of MCC, no patients with microscopic metastases developed nodal relapse, whereas regional recurrence-free survival rates at 2 years for patients with macroscopic metastatic nodal disease were 73% to 78%, depending on treatment.[23] These data would suggest that regional control can be successfully achieved when the nodal burden of disease is low, noting the well documented benefits of RT,[19] and thus we recommend performing SLNB in patients with clinically node negative disease and treating the nodal basin using RT if the SLNB is positive. Patients identified as SLNB negative are placed on a close observation policy noting the 10% to 20% risk of a false negative SLNB.[10]

None of our patients received adjuvant chemotherapy, and given the rate of distant relapse, there remains a need for effective systemic therapy in the adjuvant setting in patients with microscopic and macroscopic nodal metastasis, but MCC is a disease largely of elderly adults, for whom chemotherapy-related toxicity is not an insignificant concern. There has been a multitude of systemic treatments investigated, but none have proven efficacious with acceptable toxicity.[24]

Conclusion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusion
  7. References

Although our study involved a small number of patients, the difference in DFS and OS was nonsignificant between patients with microscopic nodal metastases and those with macroscopic nodal metastases, which suggests that even the presence of microscopic nodal metastases can have a negative effect on outcome. Given that more than half of our patients with metastatic nodal metastasis relapsed, often distant, there remains an ongoing need to develop an effective systemic treatment for this disease.

References

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusion
  7. References