We read with interest the article ‘‘Prognostic value of oncoprotein expressions in thyroid papillary carcinoma’’ by A. Z. Balta, A. I. Filiz, Y. Kurt, I. Sucullu, E. Yucel and M. L. Akin [1] because our most recent studies of p53 protein have also been showing that immunohistochemical (IHC) staining is augmented in patients with papillary thyroid carcinoma (PTC). However, we found some different associations that deserve further reflection, especially in what concerns clinical pathological aspects of PTC. We analyzed a large series of cases: 206 patients with differentiated thyroid cancer (DTC), including 164 PTC (106 classic form; 54 follicular variant; and 04 tall cell variant) and 42 follicular carcinomas, using monoclonal mouse anti-human p53 antibody (clone DO-7; monoclonal, DAKO, Carpenteria, CA, USA). In addition, we studied 105 benign thyroid tissues including: 50 nodular goiters, 55 follicular adenomas and 18 normal thyroid tissues. Our patients were all managed according to a standard protocol and followed up for 53.8 ± 41.0 months. Although our quantitative analysis using the automated cellular imaging system (ACIS-III) (Chroma Vision Medical Systems, Inc, DAKO) identified mean values of 5.11 % and medians of 1.71 % of p53 nuclei expression, we concur with Balta et al. interpretation and considered all the benign cases negative. Conversely, 28.67 % of our malignant tissues expressed p53 with a mean of 41.26 % and a median of 28.67 % stained nuclei. Balta et al. observed that p53 expression was significantly associated with the presence of lymph node metastasis and extrathyroidal invasion (p = 0.003 and p = 0.004, respectively). In contrast, we found higher expression of p53 in patients presenting better prognostic features, such as: females (p = 0.0367); smaller (\2 cm) and solitary tumors (p = 0.0105) that did not present extrathyroidal invasion (p = 0.0241). P53 was also more expressed in patients evolving free of disease than in patients with persistence or recurrence (p = 0.0310). A proportional hazard regression analysis (Cox’s multivariate proportional hazard model) identified age (p = 0.0070; HR = 1.0675; CI 95 %, 1.0180–1.1195) and the presence of metastasis at diagnosis (p = 0.0384; HR = 4.6889; CI 95 %, 1.0862–20.2417) as independent variables that influence relapse-free survival, excluding any role of p53. In addition, the Kaplan–Meier survival curve did not identify p53 as a factor affecting recurrence-free survival. The differences between our findings and those by Balta et al. [2] could be related to the patients investigated, since they have selected a group that does not correspond to the usual profile of DTC patients seen in clinical practice. In fact, they described a study population of 53 men and 34 women, whereas it is well known that 70–80 % of the DTC patients are females. Male patients are usually diagnosed in more advanced stages and may present more aggressive features and a worse outcome than females. In addition, Balta et al. classified 42 out of their 47 PTC patients as high-risk patients and only 5 as low-risk patients. Our study population (80.4 % women and 19.6 % men), with a large majority of low-risk patients (129 out of 206 DTC cases— 62.6 % were pTNM stages I and II), is more representative of the DTC usual patients. Although our results contrast with the ones described by Balta et al., they evidence the importance of clinical and M. A. Marcello (&) E. C. Morari L. S. Ward University of Campinas (FCM-Unicamp), Campinas, Sao Paulo, Brazil e-mail: marjoryam@gmail.com
Read full abstract