Abstract
Objective
To assess the role of lymph nodes micrometastases in laryngeal squamous cell carcinoma and correlate the results with survival.
Methods
We performed immunohistochemical analyses of lymph nodes after the resection of 126 patients for detection of micrometastasis. The lymph nodes were examined with hematoxylin and eosin (HE)–stained and cytokeratin (CK) antibodies AE1/AE3 stained. Recurrences and metastases were recorded during follow-up. The Kaplan-Meier method was used for survival analysis.
Results
In total, 126 patients underwent neck dissection. Forty-one patients were HE positive (group 1), while 85 were HE negative. Thirty-three of these HE negative patients were CK positive (group 2), while 52 were CK negative (group 3). Patients in groups 2 and 3 had a different outcome ( P < .001). Survival was worse in patients in group 2 (10-year survival of 52.12% vs 81.16% in group 3, P < .01).
Conclusion
Immunohistochemical analysis is an efficient way to detect micrometastasis in lymph nodes after the resection of conventionally node-negative patients. The detection of CK-positive cells is an independent prognostic factor, and more aggressive treatment should be indicated in these patients.
1
Introduction
The status of regional lymph nodes is the most important prognostic indicator in patients with laryngeal squamous cell carcinoma (LSCC). Patient survival has been shown to correlate most strongly with the TNM stage of the tumor at the time of clinical presentation. Although patients with head and neck squamous cell carcinoma without cervical lymph node involvement have a better prognosis than those with node metastasis, up to 40% will develop recurrent disease and distant metastasis. This phenomenon supports a hypothesis that some tumor cells (especially in lymph nodes) are not detected by conventional histopathology.
Conventional assessment of lymph nodes by hematoxylin and eosin (HE) staining is performed by one to 2 slices of each lymph node. Since routine histopathological examinations only detect metastases larger than 2 mm, micrometastatic cell clusters are not found during conventional HE staining . Immunohistochemical and molecular examinations facilitate the detection of single tumor cells or micrometastases smaller than 2 mm as defined by Hermanek et al. . Cytokeratine antibodies are widely used to detect and differentiate small epithelial tumor cell clusters with sufficient tissue contrast. Cytokeratines belong to a family of water-soluble proteins forming the cytoskeleton of epithelial cells . Since healthy lymph nodes do not contain epithelial cells, CK positivity in lymph nodes in patients with laryngeal cancer may indicate the presence of tumor cells.
In various tumors, several studies have assessed the prevalence and prognostic value of lymphatic micrometastasis in patients without evidence of nodal involvement on routine histopathology. In all of these studies, the lymph node classification was upstaged when examined by serial sectioning, immunohistochemistry or reverse transcriptase–polymerase chain reaction (RT-PCR) analysis. However, the influence of lymphatic micrometastasis on prognosis has not been clearly established in LSCC, while some studies have reported that micrometastasis can adversely affect outcome , especially in non–small cell lung carcinoma and in esophageal cancer .
The metastatic status of lymph nodes established on the basis of routine histopathologic examination is a decisive factor to perform adjuvant therapy after surgery for LSCC. It may be relevant to establish the prognostic significance of detection of micrometastasis or occult cancer cells in patients with LSCC and negative cervical lymph nodes by routine histopathology, as adjuvant therapy may be indicated in these patients.
In this study, we performed immunohistochemical analyses by CK AE1/AE3 of lymph nodes after resection of LSCC patients for detection of micrometastasis. In order to assess the prognostic significance of micrometastases, local recurrences, neck recurrences, and metastases were recorded during follow-up and the Kaplan-Meier method was used for survival analysis.
2
Methods
2.1
Patients
From January 1996 to June 2009, a total of 126 patients with LSCC underwent laryngeal resection and selective neck dissection in our department. All patients had at least cT1-tumor; none had evident distant metastases. Staging routinely included laryngoscope and computed tomography. Patients with locally advanced laryngeal cancer who received neoadjuvant chemoradiotherapy were excluded because of stage migration before surgical treatment. Patients with positive surgical safety margin were excluded as they were received more aggressive treatment (reoperation or postoperative radiochemotherapy).
The median age of the 126 patients (90% men) was 60 years (range, 33–79). The location of cancer was in the glottis (n = 67, 53.2%), supraglottis (n = 56, 44.4%), subglottis (n = 3, 2.4%). After resection, the final T stages were as follows: pT1 (n = 15, 11.9%); pT2 (n = 56, 44.4%); pT3 (n = 55, 43.7%). None of the patients had a pT4 stage.
2.2
Hematoxylin and eosin and immunohistochemical staining
Routine postoperative pathology HE staining was performed in all 126 patients during 153 selective neck dissections (27 patients had bilateral neck dissections), which revealed negative lymph nodes (pN0) in 85 of the 126 operated patients.
For the purpose of our study, we performed immunohistochemical analyses in HE-negative nodes for detection of micrometastasis. From these 85 pN0 patients, 1543 lymph nodes were collected. The median number of assessed lymph nodes was 18.2 nodes per patient (range, 8–36).The lymph nodes were examined with 6 consecutive sections. Every first, third, and fifth section was HE-stained; every second, fourth, and sixth was stained immunohistochemically with a cytokeratin antibody AE1/AE3 (prediluted; ABCAM, Hong Kong). This antibody reacts with human epithelial cells like adenocarcinoma and squamous cell carcinoma cells without any known exception.
To confirm CK positivity in lymph nodes in patients with LSCC, we examined lymph nodes in the 41 lymph nodes positive patients. The lymph nodes were examined with 2 sections. One section was for HE staining; the other was for CK staining.
All slides were reviewed by 2 independent experienced pathologists who were unaware of the clinical data. False-positive nonneoplastic hematopoeitic cells (eg, reticular cells and plasma cells which can also show staining for cytokeratins) were discriminated from immunohistochemics on the basis of histopathologic features.
2.3
Postoperative treatment
According to the routine HE staining, patients with pN+ received a postoperative radiotherapy; 50 Gy was applied for radiation (2 Gy/day, days 1–5,weeks 1–5). In addition, patients pT3 and pN+ at the same time received Cisplatin (100mg/m 2 body surface; day 1, weeks 1 and 4) and Taxotere (100mg/m 2 body surface; day 1, weeks 1 and 4) during postoperative radiotherapy. None of the additional postoperative treatment was performed in the patients with HE staining pN−.
2.4
Follow-up
The 126 patients were followed up prospectively for 24–184 months (mean, 75.6 months) according to Société Française d’Oto-Rhino-Laryngologie et de Chirurgie de la Face et du Cou practice guidelines . Local and neck recurrences, and metastases were recorded during follow-up.
2.5
Statistical analysis
Categorical variables were compared using the χ 2 and Fisher exact tests. P values were 2 sided. Differences were considered significant for P < .05. The Kaplan-Meier method was used for survival analysis. Statistical analyses were performed using SAS version 8.02 software.
2
Methods
2.1
Patients
From January 1996 to June 2009, a total of 126 patients with LSCC underwent laryngeal resection and selective neck dissection in our department. All patients had at least cT1-tumor; none had evident distant metastases. Staging routinely included laryngoscope and computed tomography. Patients with locally advanced laryngeal cancer who received neoadjuvant chemoradiotherapy were excluded because of stage migration before surgical treatment. Patients with positive surgical safety margin were excluded as they were received more aggressive treatment (reoperation or postoperative radiochemotherapy).
The median age of the 126 patients (90% men) was 60 years (range, 33–79). The location of cancer was in the glottis (n = 67, 53.2%), supraglottis (n = 56, 44.4%), subglottis (n = 3, 2.4%). After resection, the final T stages were as follows: pT1 (n = 15, 11.9%); pT2 (n = 56, 44.4%); pT3 (n = 55, 43.7%). None of the patients had a pT4 stage.
2.2
Hematoxylin and eosin and immunohistochemical staining
Routine postoperative pathology HE staining was performed in all 126 patients during 153 selective neck dissections (27 patients had bilateral neck dissections), which revealed negative lymph nodes (pN0) in 85 of the 126 operated patients.
For the purpose of our study, we performed immunohistochemical analyses in HE-negative nodes for detection of micrometastasis. From these 85 pN0 patients, 1543 lymph nodes were collected. The median number of assessed lymph nodes was 18.2 nodes per patient (range, 8–36).The lymph nodes were examined with 6 consecutive sections. Every first, third, and fifth section was HE-stained; every second, fourth, and sixth was stained immunohistochemically with a cytokeratin antibody AE1/AE3 (prediluted; ABCAM, Hong Kong). This antibody reacts with human epithelial cells like adenocarcinoma and squamous cell carcinoma cells without any known exception.
To confirm CK positivity in lymph nodes in patients with LSCC, we examined lymph nodes in the 41 lymph nodes positive patients. The lymph nodes were examined with 2 sections. One section was for HE staining; the other was for CK staining.
All slides were reviewed by 2 independent experienced pathologists who were unaware of the clinical data. False-positive nonneoplastic hematopoeitic cells (eg, reticular cells and plasma cells which can also show staining for cytokeratins) were discriminated from immunohistochemics on the basis of histopathologic features.
2.3
Postoperative treatment
According to the routine HE staining, patients with pN+ received a postoperative radiotherapy; 50 Gy was applied for radiation (2 Gy/day, days 1–5,weeks 1–5). In addition, patients pT3 and pN+ at the same time received Cisplatin (100mg/m 2 body surface; day 1, weeks 1 and 4) and Taxotere (100mg/m 2 body surface; day 1, weeks 1 and 4) during postoperative radiotherapy. None of the additional postoperative treatment was performed in the patients with HE staining pN−.
2.4
Follow-up
The 126 patients were followed up prospectively for 24–184 months (mean, 75.6 months) according to Société Française d’Oto-Rhino-Laryngologie et de Chirurgie de la Face et du Cou practice guidelines . Local and neck recurrences, and metastases were recorded during follow-up.
2.5
Statistical analysis
Categorical variables were compared using the χ 2 and Fisher exact tests. P values were 2 sided. Differences were considered significant for P < .05. The Kaplan-Meier method was used for survival analysis. Statistical analyses were performed using SAS version 8.02 software.
3
Results
3.1
Hematoxylin and eosin and CK positivity
No epithelial cells with atypical nuclear features characteristic of carcinoma cells were identified in lymph nodes, although occasional staining for keratin was observed in stellate cells located in the interfollicular regions of the nodes consistent with staining of reticulum cells. However, this pattern was easily distinguishable from the staining of carcinoma cells. The optical microscopy results were shown in Fig. 1 .
