Tumor deposits in head and neck carcinomas

Transcription

1 ORIGINAL ARTICLE Tumor deposits in head and neck carcinomas Sulen Sarioglu, MD, 1 * Nilhan Akbulut, MD, 1 Selen Iplikci, MD, 1 Barbaros Aydin, MD, 2 Ersoy Dogan, MD, 3 Mehtat Unlu, MD, 1 Hulya Ellidokuz, MD, 4 Emel Ada, MD, 5 Fadime Akman, MD, 2 Ahmet Omer Ikiz, MD, PhD 3 1 Department of Pathology, Dokuz Eylul University, Faculty of Medicine, Izmir, Turkey, 2 Department of Radiation Oncology, Dokuz Eylul University Faculty of Medicine, Izmir, Turkey, 3 Department of Otorhinolaryngology Head and Neck Surgery, Dokuz Eylul University, Faculty of Medicine, Izmir, Turkey, 4 Department of Preventive Oncology, Dokuz Eylul University, Institute of Oncology, Izmir, Turkey, 5 Department of Radiodiagnostics, Dokuz Eylul University, Faculty of Medicine, Izmir, Turkey. Accepted 18 December 2014 Published online 21 August 2015 in Wiley Online Library (wileyonlinelibrary.com). DOI /hed ABSTRACT: Background. Tumor deposits, nodules in the peritumoral adipose tissue with no architectural residue of lymph node, have previously been described in colorectal adenocarcinomas. To date, however, there has been no examination of tumor deposits in head and neck squamous cell carcinoma (HNSCC). Methods. Neck dissection specimens of 140 patients with HNSCC were reevaluated for tumor deposits. Results. Tumor deposits were detected in 24 cases (17%). Cases with tumor deposits had more lymphatic invasion (p 5.007), higher pathological N classification (p 5.00), and more frequently showed distant metastasis (p 5.003). Disease-free and overall survival were significantly shorter for tumor deposit positive cases (p and p 5.005, respectively). Only tumor deposits were significant for overall survival. Tumor deposits increased the risk of recurrent disease 2294 times. Tumor deposits and pericapsular invasion were identified as independent prognostic markers; tumor deposits increased the risk of death from disease 3.4 times, whereas pericapsular invasion was associated with a 2.2-fold increase in the risk of death. Conclusion. These results highlight the existence of tumor deposits in neck dissection specimens of HNSCC and their association with poor prognosis. VC 2015 Wiley Periodicals, Inc. Head Neck 38: E256 E260, 2016 KEY WORDS: head and neck, squamous cell carcinoma, tumor deposit, neck dissection, prognosis *Corresponding author: S. Sarioglu, Department of Pathology, Dokuz Eylul University, Faculty of Medicine, Izmir, Turkey. sulen.sarioglu@deu.edu. tr or sulensari@gmail.com INTRODUCTION Tumor deposits were first described by Gabriel et al 1 in 1935 in the context of colorectal adenocarcinomas. Seventyfive years later, in 2010, they were included in the Union for International Cancer Control classification for colorectal carcinomas. 2 This specific mechanism of tumor spread has been best characterized in colorectal cancers; however, even in colorectal cancers, proper use of tumor deposits as prognostic markers remains controversial Tumor deposits have also been identified in gastric carcinomas, as well as pancreatic carcinomas, and cholangiocarcinomas. 10 Among squamous cell carcinomas (SCCs), only esophageal SCC has previously been documented as exhibiting tumor deposits, and only in a single published study. 21 Tumor deposits are described as irregular discrete tumor masses in pericolorectal fat, discontinuous from the leading edge of the tumor. For the diagnosis of tumor deposits, no residual lymph node tissue should be observed, and the tumor deposits should be within the lymphatic drainage of the primary tumor. 22 Tumor deposits may have been observed by many pathologists in neck dissection specimens from patients with head and neck squamous cell carcinomas (HNSCCs), and may have been diagnosed as metastatic lymph nodes and/or free tumor deposits. This study came about in response to the sparse information about tumor deposits in SCC, particularly HNSCC. The main purpose of this study was to evaluate the presence of tumor deposits in a series of HNSCC samples in order to estimate their incidence, as well as their association with histopathological and clinical prognostic features and survival. MATERIALS AND METHODS This study was approved by the Dokuz Eylul University Ethics Committee. Two hundred patients with laryngeal, hypopharyngeal, oropharyngeal, and oral cavity SCC were diagnosed and treated at Dokuz Eylul University Hospital between 1986 and 2012, but only 140 were selected for this study as only these cases had neck dissection materials and complete follow-up information. None of the patients had received chemo/radiotherapy before surgery. Age, sex, tumor location, clinical and pathological tumors, lymph node classification, pericapsular invasion, tumor grade, treatment scheme with chemo/radiotherapy, time and localization of locoregional recurrence, and distant metastasis were determined from the medical records of the patients. Histopathological evaluation Previously, the neck specimens were examined and sampled, as described in the Cancer Protocols of the E256 HEAD & NECK DOI /HED APRIL 2016

2 TUMOR DEPOSITS FIGURE 1. (A) Tumor deposit with irregular contours (hematoxylin-eosin stain; original magnification 32). (B) Tumor deposit with irregular contours (hematoxylin-eosin stain; original magnification 34). (C) Tumor deposit with regular contours but no lymph node structure is observed (hematoxylin-eosin stain; original magnification 32). (D) Higher magnification of the tumor deposit shown in panel C (hematoxylin-eosin stain; original magnification 340). College of American Pathologists for Head and Neck carcinomas,23 and any lesion suspicious for metastatic disease was sampled for histopathological examination. During the present study, the sections from neck dissection materials were reevaluated for lymph nodes and tumor deposits. Any tumor mass, either circumscribed or with irregular contours, devoid of lymph node architecture was identified as a tumor deposit (Figure 1A 1D). The total number of lymph nodes found to contain tumor cells, metastatic lymph nodes, and lymph nodes with pericapsular invasion were also reevaluated; based on these results, the N classification was updated for each case. Statistical analysis Cases were grouped according to the presence or absence of tumor deposits. After separation, statistical analysis was conducted by Scientific Package for Social Sciences. For comparison of the groups, nonparametric tests, such as a chi-square test and Mann Whitney U test, were applied. Survival curves were plotted by the Kaplan Meier method. For comparing survivals of different groups, the log-rank test was applied. Multivariate analysis by the Cox proportional hazards model was performed for overall survival. RESULTS The study population consisted of 121 men (86.42%) and 19 women (13.58%). The mean age of the patients was years; median age of patients was years (range, years.) Eighty-four larynx (60.00%), 31 oral cavity (22.14%), 22 hypopharynx (15.71%), and 3 tonsilla palatina (0.37%) SCC cases were included in this study. One hundred thirty-six patients (97.14%) had classical SCC and 4 patients (2.86%) had basaloid SCC. Tumors were well differentiated in 50 cases (35.71%), whereas 58 cases (41.43%) were moderately differentiated, and 32 cases (22.86%) were poorly differentiated. Pathological T classifications were as follows: T1 for 9 patients (6.42%), T2 for 20 patients (14.29%), T3 for 48 patients (34.29%), and T4 for 63 patients (45.00%). Fiftysix patients (40.00%) had no lymph node metastasis (pn0), and the numbers of pn1, pn2a, pn2b, pn2c, and pn3 cases were 31 (22.14%), 11 (7.86%), 31 (22.14%), 9 (6.43%), and 2 (1.43%), respectively. Twenty-one patients (15%) had pericapsular invasion. Twenty-four patients (17.14%) had tumor deposits. Among these cases, 12 patients (50%) had laryngeal carcinoma, 7 patients (29.2%) had oral carcinoma, and the remaining 5 patients (20.8%) had hypopharyngeal carcinoma. The number of tumor deposits ranged from 1 to 14 ( ) and measured 3 to 24 mm ( ). Fifty-eight patients (41.43%) were treated by only radical surgery and 80 patients (57.41%) received additional chemo/radiotherapy. Two patients discontinued treatment during chemotherapy and radiotherapy sessions. The HEAD & NECK DOI /HED APRIL 2016 E257

3 SARIOGLU ET AL. TABLE 1. Clinicopathologic features of head and neck carcinoma with and without tumor deposits. Negative Tumor deposit Positive Total Sex Men 99 (81.8) 22 (18.2) 121 (86.4) Women 17 (89.5) 2 (10.5) 19 (13.6) Tumor localization Larynx 72 (85.7) 12 (14.3) 84 (60) Oral cavity 24 (77.4) 7 (22.6) 31 (43.6) Tonsil 3 (100) 0 3 (2.1) Hypopharynx 17 (77.27) 5 (22.72) 22 (15.7) Histological type Epidermoid 113 (83.1) 23 (16.9) 136 (97.2) Basosquamous 3 (75) 1 (25) 4 (2.8) Grade Well 43 (87.76) 6 (12.25) 49 (35) Moderately 46 (80.7) 11 (19.3) 57 (40.7) Poorly 26 (81.25) 6 (18.75) 32 (22.86) Clinical stage I 1 (100) 0 (0) 1 (0.71) II 12 (100) 0 (0) 12 (8.57) III 42 (97.67) 1 (2.33) 43 (30.71) IV 61 (72.62) 23 (27.38) 84 (57) Clinical T classification ct1 5 (83.33) 1 (16.67) 6 (4.29) ct2 25 (83.33) 5 (16.67) 30 (21.43) ct3 47 (81.03) 11 (18.97) 58 (41.43) ct4 39 (84.78) 7 (15.22) 46 (32.86) Clinical N classification cn0 70 (95.89) 3 (4.11) 73 (51.05) cn1 18 (100) 0 (0) 18 (12.86) cn2a 20 (58.82) 14 (41.18) 34 (24.29) cn2b 5 (50.00) 5 (50.00) 10 (7.14) cn2c 3 (75.00) 1 (25.00) 4 (2.86) cn3 0 (0) 1 (100) 1 (0.71) Pathologic stage I 1 (100) 0 (0) 1 (0.71) II 8 (100) 0 (0) 8 (5.71) III 28 (96.55) 1 (3.45) 29 (20.71) IV 79 (77.45) 23 (22.55) 102 (72.86) Pathologic T classification pt1 7 (77.78) 2 (22.22) 9 (6.43) pt2 16 (80) 4 (20) 20 (14.29) pt3 42 (85.71) 7 (14.29) 49 (35) pt4 51 (82.26) 11 (17.74) 62 (44.29) Pathologic N classification pn0 55 (98.21) 1 (1.79) 56 (40) pn1 27 (87.10) 4 (12.90) 31 (22.14) pn2a 9 (81.89) 2 (18.18) 11 (7.86) pn2b 19 (61.29) 12 (38.71) 31 (22.14) pn2c 6 (66.67) 3 (33.33) 9 (6.43) pn3 0 (0) 2 (100) 2 (1.43) Surgical margin Negative 98 (81.67) 22 (18.33) 120 (85.71) Positive 18 (90) 2 (10) 20 (14.29) Pericapsular invasion Negative 95 (79.83) 24 (20.17) 119 (85) Positive 21 (100) 0 (0) 21 (15) Perineural invasion Negative 108 (83.08) 22 (16.92) 130 (92.86) Positive 8 (80) 2 (20) 10 (7.14) Vascular invasion Negative 113 (84.33) 21 (15.67) 134 (95.71) Positive 3 (50) 3 (50) 6 (4.29) TABLE 1. Continued Negative Tumor deposit Positive Total Lymphatic invasion Negative 87 (88.78) 11 (11.22) 98 (70) Positive 29 (69.05) 13 (30.95) 42 (30) Locoregional disease Negative 95 (81.90) 21 (18.10) 116 (82.86) Positive 18 (85.71) 3 (14.29) 21 (15) Recurrence Negative 84 (86.60) 13 (13.40) 97 (69.29) Positive 29 (72.50) 11 (27.50) 40 (28.57) Total 116 (82.86) 24 (17.14) 140 (100) Cases with tumor deposits were more likely to exhibit lymphovascular invasion (p 5.007), higher pathological N classification (p 5.00) and distant recurrence (p 5.003) and less pericapsular invasion (p 5.02). mean follow-up period was months (range, months), with a median follow-up time of 47 months. Forty patients were diagnosed with local recurrences and/or distant metastases during follow-up. Locoregional disease was seen in 18 patients (12.86%), whereas 6 patients (4.29%) had only local recurrences. Twenty-two patients (15.71%) developed distant metastatic disease during follow-up. Eighty-nine patients (63.57%) are still alive, whereas 51 patients (36.43%) died. Thirteen patients (9.29%) died of primary head and neck carcinoma, 7 patients (5%) died of second primary tumor, 16 patients (11.43%) died of metastatic disease, and 15 patients (10.71%) died of other causes. There was no difference for tumor deposits when sex, histological type, tumor grade, tumor localization, clinical and pt classifications, surgical margin, perineural invasion, and number of metastatic lymph nodes were considered (p 5.52; p 5.53; p 5.58; p 5.52; p 5.52; p 5.68; p 5.75; and p 5.87, respectively). Cases with tumor deposits had more lymphovascular invasion (p 5.007), higher pathological N classification (p 5.00) and distant recurrence (p 5.003), and less pericapsular invasion (p 5.02; Table 1). Disease-free survival rates at 12 and 24 months were 93.7% and 85.2%, respectively, for cases without tumor deposits. In cases with tumor deposits, survival rates were 64.1% and 37.9% at 12 and 24 months, respectively, a significant decrease in survival compared to cases without tumor deposits (p 5.016). Overall survival rates at 12 and 24 months were 94.2% and 89% for tumor deposit negative cases, whereas in tumor deposit positive cases, survival rates were lower at 56.8% and 31.6% at 12 and 24 months, respectively (p 5.05). Mean estimated recurrence-free survival time for all cases, tumor deposit negative, and tumor deposit positive cases were: , , and months, respectively (p 5.005). Mean estimated survival time for these groups was , , and months, respectively (p 5.016). Relationships of tumor deposits with disease-free and overall survival are shown in Figures 2 and 3. E258 HEAD & NECK DOI /HED APRIL 2016

4 TUMOR DEPOSITS TABLE 2. Cox regression analysis for disease-free survival. Hazards ratio (95% confidence interval) p value Step 5 Tumor deposits ( ).02 pt, pn, pericapsular invasion, and surgical margin positivity, as well as tumor deposits, were included. Only tumor deposits attained statistical significance, with the relative risk of recurrent disease increased 2.29 times (range, ; p 5.02). of death of disease 3.4-fold, whereas pericapsular invasion increased the risk of death 2.2-fold (Tables 2 and 3). FIGURE 2. Disease-free survival for cases with and without tumor deposits. Multivariate analysis was performed based on wellrecognized prognostic factors, including pt, pn, pericapsular invasion, and surgical margin positivity, as well as tumor deposits both for disease-free and overall survival. Of all clinicopathological prognostic markers examined, only the presence of tumor deposits showed a statistically significant link to disease-free and overall survival. Furthermore, tumor deposit positivity increased the risk of recurrent disease 2.29-fold (range, ), and was the only prognostic indicator significantly linked to the risk of recurrent disease (p 5.02). Tumor deposits and pericapsular invasion were identified as independent prognostic markers for overall survival. When analyzed by Cox regression, tumor deposit positivity increased the risk DISCUSSION Tumor deposits, which have previously been described in cancers of the gastrointestinal regions, can also be found in cancers of the head and neck region. The mechanism and clinical significance of these masses remain under investigation. The evaluation of diagnostic categorization of tumor deposits in colorectal carcinomas is summarized by Nagtegaal et al. 14 According to the latest classification, cases with tumor deposits without any other lymph node metastases are pn1c; if there are any metastatic lymph nodes, tumor deposit positivity is reported separately. The tumor spread in order to form a deposit away from the primary tumor seems to be either by intralymphatic or perilymphatic, perivascular and or perineural pathways, and, in most cases, this is the result of more than 1 pathway. 8 This may resemble a journey along an allocated highway, allowing escape from immune system surveillance. At some point during their spread via these routes, the tumor cells home to a specific destination, where they develop discrete masses. Although this destination is frequently a lymph node, this does not seem to be the case in instances of free tumor deposits. 5 7 One of the explanations for such a deposit may be lymph node metastasis with extensive pericapsular invasion spread obliterating the whole lymph node architecture. The morphology of some tumor deposits, however, is not reminiscent of such a mechanism. Two types of deposits are observed: those with and without regular contours. The former is likely related to lymphatic/perilymphatic spread, whereas the latter is likely related to vascular/perivascular spread. 5 In previous TNM classification schemes for colorectal carcinomas, tumor deposits with regular contours were accepted as lymph node metastases, whereas those with irregular borders were accepted as vascular invasion. 8 Clinical and histopathological findings led to their TABLE 3. Cox regression analysis for overall survival. Hazards ratio (95% confidence interval) p value Step 4 Tumor deposits ( ).003 Pericapsular invasion ( ).06 FIGURE 3. Overall survival for cases with and without tumor deposits. pt, pn, pericapsular invasion, surgical margin positivity, and the presence of tumor deposits were included. Of these markers, only pericapsular invasion and tumor deposits showed statistically significant links to the risk of disease-related death. Pericapsular invasion increased the risk times (range, ; p 5.06), and tumor deposits increased the risk of death times (range, ; p 5.003). HEAD & NECK DOI /HED APRIL 2016 E259

5 SARIOGLU ET AL. description as tumor deposits The controversy is obvious, as it is difficult to designate a tumor mass as a metastatic lymph node when no lymph node structure can be observed, and equally difficult to designate a mass as vascular invasion without observing a vessel. It may also be difficult to differentiate tumor deposits from tumor remnants observed after chemo/radiotherapy. 24 In this study, however, none of the patients had received any chemo/radiotherapy before surgery. Ueno et al 5 had previously observed tumor deposits in 16.6% of colon and 23.3% of rectal carcinomas. In a series of 228 colorectal advanced carcinoma cases, Puppa et al 6 identified tumor deposits in 43.85% of the cases. Tumor deposits were found in 17.8%, 23.9%, and 24% of cases in different studies of gastric carcinomas If this is the case for adenocarcinomas of the gastrointestinal tract, it should perhaps come as no surprise that tumor deposits are also observed in HNSCCs. Like tumor budding, a histopathological feature first described in colorectal tumors that seems to be a poor prognostic marker when present in head and neck carcinomas, tumor deposits may possess utility as a novel prognostic marker in head and neck carcinomas. In this series, we have observed tumor deposits in 17.14% of head and neck carcinoma cases. The results suggest that oral cavity and hypopharyngeal carcinoma cases are more likely to show tumor deposits than laryngeal carcinomas, although this trend was not found to be statistically significant. We did not find statistical significance for the link between tumor deposits and clinical and pathological T classification. However, tumor deposits were observed more frequently in cases with lymphatic invasion and lymph node metastasis. The disease-free and overall survival times were significantly shorter for cases with tumor deposits, and recurrent disease was more frequent in cases with tumor deposits. The clinicopathological prognostic factors other than tumor deposits were not statistically significant by univariate analysis in this series. This may be due to wellknown behavior differences between primary tumors at different head and neck regions. By multivariate analysis, however, tumor deposits were found to show a significant link to disease-free and overall survival. Tumor deposits were previously reported as either metastatic lymph nodes or free tumor deposits. This series provides information about the existence of tumor deposits in head and neck carcinomas, as well as their observed association with poor prognosis. However, as our series was relatively small, further studies conducted with more patients are still required in order to reveal the actual impact of tumor deposits on prognosis of head and neck carcinomas. REFERENCES 1. Gabriel WB, Dukes C, Bussey HJR. Lymphatic spread in cancer of the rectum. Br J Surg 1935;23: Sobin LH, Gospodarowicz M, Wittekind C, editors. International Union Against Cancer TNM Classification of Malignant Tumours. 7th ed. Hoboken, NJ: Wiley Blackwell; Ueno H, Mochizuki H. Clinical significance of extrabowel skipped cancer infiltration in rectal cancer. Surg Today 1997;27: Goldstein NS, Turner JR. Pericolonic tumor deposits in patients with T3N1MO colon adenocarcinomas: markers of reduced disease free survival and intra-abdominal metastases and their implications for TNM classification. Cancer 2000;88: Ueno H, Mochizuki H, Hashiguchi Y, et al. Extramural cancer deposits without nodal structure in colorectal cancer: optimal categorization for prognostic staging. Am J Clin Pathol 2007;127: Puppa G, Maisonneuve P, Sonzogni A, et al. Pathological assessment of pericolonic tumor deposits in advanced colonic carcinoma: relevance to prognosis and tumor staging. Mod Pathol 2007;20: Ueno H, Mochizuki H, Shirouzu K, et al. Multicenter study for optimal categorization of extramural tumor deposits for colorectal cancer staging. Ann Surg 2012;255: Nagtegaal ID, Quirke P. Colorectal tumour deposits in the mesorectum and pericolon; a critical review. Histopathology 2007;51: Lo DS, Pollett A, Siu LL, Gallinger S, Burkes RL. Prognostic significance of mesenteric tumor nodules in patients with stage III colorectal cancer. Cancer 2008;112: Puppa G, Ueno H, Kayahara M, et al. Tumor deposits are encountered in advanced colorectal cancer and other adenocarcinomas: an expanded classification with implications for colorectal cancer staging system including a unifying concept of in-transit metastases. Mod Pathol 2009;22: Shimada Y, Takii Y. Clinical impact of mesorectal extranodal cancer tissue in rectal cancer: detailed pathological assessment using whole-mount sections. Dis Colon Rectum 2010;53: Belt EJ, van Stijn MF, Bril H, et al. Lymph node negative colorectal cancers with isolated tumor deposits should be classified and treated as stage III. Ann Surg Oncol 2010;17: Al Sahaf O, Myers E, Jawad M, Browne TJ, Winter DC, Redmond HP. The prognostic significance of extramural deposits and extracapsular lymph node invasion in colon cancer. Dis Colon Rectum 2011;54: Nagtegaal ID, Tot T, Jayne DG, et al. Lymph nodes, tumor deposits, and TNM: are we getting better? J Clin Oncol 2011;29: Greene FL. Tumor deposits in colorectal cancer: a moving target. Ann Surg 2012;255: Tong LL, Gao P, Wang ZN, et al. Is the seventh edition of the UICC/AJCC TNM staging system reasonable for patients with tumor deposits in colorectal cancer? Ann Surg 2012;255: Nagayoshi K, Ueki T, Nishioka Y, et al. Tumor deposit is a poor prognostic indicator for patients who have stage II and III colorectal cancer with fewer than 4 lymph node metastases but not for those with 4 or more. Dis Colon Rectum 2014;57: Sun Z, Wang ZN, Xu YY, et al. Prognostic significance of tumor deposits in gastric cancer patients who underwent radical surgery. Surgery 2012; 151: Lee HS, Lee HE, Yang HK, Kim WH. Perigastric tumor deposits in primary gastric cancer: implications for patient prognosis and staging. Ann Surg Oncol 2013;20: Ersen A, Unlu MS, Akman T, et al. Tumor deposits in gastric carcinomas. Pathol Res Pract 2014;210: Sakai M, Suzuki S, Sano A, et al. Significance of lymph node capsular invasion in esophageal squamous cell carcinoma. Ann Surg Oncol 2012;19: College of American Pathologists. Available at: docs/committees/cancer/cancer_protocols/2013/colon_13protocol_3300. pdf. Accessed June 1, College of American Pathologists. Available at: docs/committees/cancer/cancer_protocols/2013/liporalcav_13protocol_ 3200.pdf. Accessed June 1, Kinoshita H, Watanabe T, Yanagisawa A, Nagawa H, Kato Y, Muto T. Pathological changes of advanced lower-rectal cancer by preoperative radiotherapy. Hepatogastroenterology 2004;51: Sarioglu S, Acara C, Akman FC, et al. Tumor budding as a prognostic marker in laryngeal carcinoma. Pathol Res Pract 2010; 206: Almangush A, Bello IO, Keski S antti H, et al. Depth of invasion, tumor budding, and worst pattern of invasion: prognostic indicators in early-stage oral tongue cancer. Head Neck 2014;36: Almangush A, Salo T, Hagstr om J, Leivo I. Tumour budding in head and neck squamous cell carcinoma a systematic review. Histopathology 2014;65: E260 HEAD & NECK DOI /HED APRIL 2016