Role of BRAF molecular analysis in the management of papillary thyroid carcinoma: analysis of cytological and histological samples.

BACKGROUND: Although fine needle aspiration (FNA) is the standard diagnostic test for the characterization of a suspicious thyroid nodule, in some cases cytological evaluation is inconclusive. The aim of this study was to determine the role of BRAF mutation in aiding diagnosis and to verify whether archival cytological samples could be suitable for molecular analysis. METHODS: Eighty-five patients with suspicious (Thy4) or follicular (Thy3) lesions on cytology were resubmitted to a second FNA for BRAF mutation analysis. Of these, 56 subsequently underwent surgery. The usefulness of archival samples for molecular analysis was also studied in a second cohort of 42 patients with a confirmed diagnosis of papillary thyroid carcinoma for whom both archived paraffin-embedded histological samples and cytological smears were available. A further 15 patients with paired fresh FNA and archived cytological and histological samples were recruited. RESULTS: BRAF mutation was found in the fresh FNA samples from 10 of 56 patients who had surgery with previous inconclusive cytology (4/45, 9%, Thy3 and 6/11, 55%, Thy4). The BRAF test showed a specificity and positive predictive value of 100% (26/26 and 10/10, respectively), sensitivity of 33% (10/30) and negative predictive value of 57% (26/46). There was absolute concordance between the BRAF results obtained with 42 histological and cytological archived samples. BRAF analysis on 15 archived cytological samples showed absolute concordance with histology, whereas there was one false negative on the matched fresh FNA. CONCLUSION: BRAF analysis is a highly specific test that can facilitate cytological diagnosis in some cases and can also be performed on archived cytological samples.

[Applicative aspects of liquid-based cytology in cervical cancer screening]. / Aspetti applicativi della citologia in fase liquida nello screening del cervico-carcinoma.

The quality of cytological services is the very heart of the prevention of cervical pathologies. Indeed, various studies have demonstrated that inadequate sampling, mistakes made in the organisational and management methods of the screening programme, and incorrect diagnoses result in unnecessarily high incidence and mortality rates. The aim of this work is to compare the effectiveness of two different methods, i.e. a conventional smear test and a liquid based ThinPrep (TP) test. Said methods were tested on a sample 453 cases diagnosed as being "Atypical Squamous Cells of Undetermined Significance"/"Atypical Glandular Cells of Undetermined Significance" according to the 1991 Bethesda System. All the women with an "Atypical Squamous Cells of Undetermined Significance "/"Atypical Glandular Cells of Undetermined Significance" cytological diagnosis were called back within 3 months for a ThinPrep test, as part of the Level 2 diagnostic controls of a cervical cancer screening programme. Of the initial diagnoses of "Atypical Squamous Cells of Undetermined Significance"/"Atypical Glandular Cells of Undetermined Significance" with a conventional smear test, 124 cases (27.4%) were classified as being adequate, while 329 (72.6%) were satisfactory, although they did have limited indicators of quality. Upon repetition of the cytology with a ThinPrep test, 322 cases (71.1%) were found to be adequate, 129 (28.4%) "suboptimal" and 2 inadequate (p < 0.0001). The main reasons for insufficient results in conventional smear tests are: bad preservation (40.2%), the presence of granulocytes (36.4%), intense phlogosis (12.1%) and erythrocytes (5.5%). In liquid based smear tests, the main indicator of quality is the absence of endocervical glandular cells (56.7%). As for the cytological diagnosis, the use of ThinPrep supplied the following results: of the 453 cases diagnosed initially as being "Atypical Squamous Cells of Undetermined Significance"/"Atypical Glandular Cells of Undetermined Significance", 371 (84.1%) were negative, 54 (11.9%) "Atypical Squamous Cells of Undetermined Significance "/"Atypical Glandular Cells of Undetermined Significance" and 18 (4%) L-SIL (p < 0.0001). Histological follow-up of the 18 cases with L-SIL confirmed the presence of a dysplastic lesion in 8 out of 12 cases (66.7%); in 4 cases there was no consistency between the cytological and histological diagnoses, and in 6 patients no biopsy had been taken. The preliminary experience of this study, although indeed carried out on a limited number of cases, appears to show that suitable training for the collection of samples in a liquid solution could improve the adequacy of the sample and thus the precision of the cytological diagnosis.

Potentiation of antiproliferative drug activity by lonidamine in hepatocellular carcinoma cells.

The ability of lonidamine (LND), a derivative of indazole-carboxylic acid, to modulate the cytotoxic activity of anticancer drugs was investigated in two human hepatocarcinoma (HCC) cell lines. The cytotoxicity of drugs used singly, in association or in sequence was evaluated using the Sulforhodamine B (SRB) assay. LND did not appreciably potentiate the effect of antitumor drugs when given before or simultaneously, in either cell line. Conversely, a synergistic interaction was observed in both cell lines when LND was given after conventional drugs. LND produced a moderate decrease in S-phase cell fraction and did not induce apoptosis. Conversely, paclitaxel (TAX) induced an important block in G2 and an increase in apoptosis. Following a 48-h TAX wash out, a progressive passage of cells from G2 to M phase was observed with a corresponding increase in apoptotic cells. Post-treatment with LND increased the cytotoxicity of some antitumor drugs, especially TAX, in hepatocarcinoma cells, possibly by preventing, as an energolytic drug, cell damage repair or by producing an additional effect on microtubule stabilization.

Restored T-cell activation mechanisms in human tumour-infiltrating lymphocytes from melanomas and colorectal carcinomas after exposure to interleukin-2.

We investigated the effects of interleukin-2 (IL-2) exposure on T-cell signal transduction molecules and apoptosis markers in tumour-infiltrating lymphocytes (TIL) isolated from 20 melanoma and 16 colorectal carcinoma metastases and expanded in vitro for therapeutic reinfusion. Before IL-2 culture, TIL showed undetectable or very low levels of T-cell receptor (TCR) epsilon chain, p56(lck), Fas ligand (FasL) and Bax expression, while Bcl-2 values were elevated. Cancer cells were characterised by low or absent Fas and Bcl-2 and high Bax expression. Notably, they also expressed FasL. After 41-48 days of IL-2 culture, TCR epsilon chain and p56(lck) expression of TIL rose to median values of approximately 80 and 30% positive cells, respectively (P<0.001), FasL expression was detected in 45% cells from melanomas (P<0.001) and in 3% from colorectal carcinomas (P=0.09), and Bax-positive cells increased from 17.5 to 70% (P=0.005). Moreover, TCR zeta chain-positive cells were significantly increased from baseline (P=0.001), Bcl-2-positive cells dropped from 50 to 1% (P=0.007) and perforin content was high, while Fas expression was not significantly modified by IL-2 culture. In conclusion, our data suggest that the degree of immunosuppression in TIL from melanomas and colorectal carcinomas is very high, and the apoptosis markers' repertoire of cancer cells resembles that of immune-privileged tissue. Interleukin-2 culture appears to restore lymphocyte activation mechanisms, resulting in consistent FasL expression and perforin production.

Quality control for biomarker determination in oncology: the experience of the Italian Network for Quality Assessment of Tumor Biomarkers (INQAT).

Biomarker analysis and evaluation in oncology is the product of a number of processes (including managerial, technical and interpretation steps) which need to be monitored and controlled to prevent and correct errors and guarantee a satisfactory level of quality. Several biomarkers have recently moved to clinical validation studies and successively to clinical practice without any definition of standard procedures and/or quality control (QC) schemes necessary to guarantee the reproducibility of the laboratory information. In Italy several national scientific societies and single researchers have activated -- often on a pilot level -- specific external quality assessment protocols, thereby potentially jeopardizing the clinical reality even further. In view of the seriousness of the problem, in 1998 the Italian Ministry of Health sponsored a National Survey Project to coordinate and standardize the procedures and to develop QC programs for the analysis of cancer biomarkers of potential clinical relevance. Twelve QC programs focused on biomarkers and concerning morphological, immunohistochemical, biochemical, molecular, and immunoenzymatic assays were coordinated and implemented. Specifically, external QC programs for the analytical phase of immunohistochemical p53, Bcl-2, c-erb-2/neu/HER2, and microvessel density determination, of morphological evaluation of tumor differentiation grade, and of molecular p53 analysis were activated for the first time within the project. Several hundreds of Italian laboratories took part in these QC programs, the results of which are available on the web site of the Network (www.cqlaboncologico.it). Financial support from the Italian Government and the National Research Council (CNR) will guarantee the pursuit of activities that will be extended to new biomarkers, to preanalytical phases of the assays, and to revision of the criteria of clinical usefulness for evaluating the cost/benefit ratio.

Role of biological markers in the clinical outcome of colon cancer.

We investigated a number of biological markers, evaluated under strict intralaboratory quality control conditions, in terms of their role in predicting clinical outcome of patients with colon cancer treated with 5-FU-containing regimens. Colon cancer tissue from 263 patients enrolled onto two randomised clinical trials were studied for their cytofluorimetrically determined DNA content and their immunohistochemically evaluated microvessel density, vascular endothelial growth factor expression, thymidylate synthase expression and tumour lymphocyte infiltration. Disease-free survival and overall survival of patients were analysed as a function of the different variables. At a median follow up of 57 months, age, gender and Dukes' stage showed an impact on disease-free survival, whereas no biological marker emerged as an indicator of better or worse disease-free survival. Only histological grade and Dukes' stage were found to influence overall survival. The different biological variables, studied with particular attention for determination reliability, proved to have no impact on the clinical outcome of patients with colon cancer. Therefore, other markers must be identified to complement clinico-pathological variables in the management of this disease.

Flow-cytometric determination of tumor cells in lymph nodes.

In solid tumors, metastasis occurs through the dissemination of tumor cells in the bloodstream and the lymphatic system. In particular, lymph node infiltration gives useful prognostic information and represents one of the most important factors for selecting the type of clinical treatment in disease management. Furthermore, the analysis of lymph node infiltration has become important for identifying patients with breast cancer or malignant melanoma who may be candidates for regional lymph node dissection. Tumor cells in lymph nodes are currently identified in tissue sections using morphological and immunohistochemical analyses, but these approaches are time-consuming, and micrometastases may escape detection. The aim of the present study was to define the potential of a flow cytometric (FCM) determination based on cell size and autofluorescence to shorten the time required for lymph node analysis. The sensitivity of the FCM approach, defined on mixtures of tumor cells from established cell lines and peripheral blood lymphocytes (PBL(s)) at different concentrations, was 1 tumor cell/1,000 PBL(s). FCM analysis was performed on 89 lymph nodes, 29 from breast, 41 from lung and 19 from colon cancer patients. Agreement between FCM and morphological results, used as gold standard, was observed in 83% of the cases, and there was a 90% sensitivity to the FCM approach for each tumor type. Disagreement was observed for 15 lymph nodes and was due, in the majority of cases (80%), to FCM-positive and morphologically negative results. A large number of patients and a more accurate pathological examination of consecutive histological sections of lymph nodes are needed to further evaluate the validity of the FCM approach.

Biomarker prediction of clinical outcome in operable breast cancer patients treated with tamoxifen.

The predictivity of tumour size, oestrogen (ER) and progesterone (PgR) receptors, 3H-thymidine labelling index (TLI), c-erbB-2 and p27kip1 expression on clinical outcome was analysed on a consecutive series of 118 postmenopausal patients with ER-positive, node-positive tumours. All patients were treated with surgery +/- radiotherapy and adjuvant tamoxifen (30 mg/day) for at least 2 years. TLI, ER, c-erbB-2 and p27kip1 were generally unrelated to each other. PgR was directly related to ER and inversely to c-erbB-2. Tumour size was inversely related to both c-erbB-2 and p27kip1 expression. At a median follow-up of 75 months, 5-year relapse-free survival was significantly lower for patients with very rapidly proliferating (HR = 2.61, 95% CI = 1.34-5.08), PgR negative (HR = 2.76, 95% CI = 1.43-5.33) or relatively low ER content (HR = 2.20, 95% CI = 1.14-4.25) tumours than for patients with tumours expressing the opposite biological profiles. Overall survival was also significantly different as a function of TLI (HR = 3.47, 95% CI = 1.52-7.93) and PgR (HR = 2.27, 95% CI = 1.00-5.15). TLI and PgR maintained an independent relevance in multivariate analysis and together were capable of identifying subgroups of patients at significantly different risk of relapse and death.

Lanreotide-induced modulation of 5-fluorouracil or mitomycin C cytotoxicity in human colon cancer cell lines: a preclinical study.

The effect on growth of the long-acting somatostatin analogue lanreotide (LAN), alone or in combination with 5-fluorouracil (5-FU) and mitomycin C (MIT), was investigated in three human colon cancer lines. Cell survival inhibition induced by LAN alone, as evaluated by sulforhodamine B assay, ranged from 20% to 40% as a function of cell line and concentration. The IC50, the concentration inhibiting cell survival by 50%, was never reached. The antiproliferative effect produced by a 48 h exposure to 5-FU or MIT was synergistically enhanced in all cell lines by a subsequent 48 h exposure to LAN. The synergistic interaction was not related to specific cell cycle perturbations or to the somatostatin receptor 2 (sst2) mRNA abundance. In conclusion, our study seems to indicate that LAN is a potentially useful modulating agent for enhancing 5-FU and MIT activity in colorectal cancer patients.

Prognostic significance of biologic markers in node-negative breast cancer patients: a prospective study.

It is generally thought that future advances in the treatment and cure of breast cancer patients will be made possible through a deeper understanding of tumor biology and an improved capability to define the prognosis of each single patient. This will lead to the formulation of new, more selective, and patient-tailored therapies. It is therefore important, when studying potential prognostic factors, to follow methodologic requirements and guidelines which involve the carrying out of prospective studies as confirmatory steps. Repeatedly or recently investigated prognostic markers (tumor size, menopausal status, ER, PgR, 3H thymidine labeling index, c-erbB-2 and p27 expression) were evaluated on a series of 286 prospectively recruited node negative breast cancer patients who underwent loco-regional treatment alone and were closely followed. The individual and relative prognostic contribution of each variable with respect to other factors, as well as their ability to identify node negative patients at risk, were assessed by univariate and multivariate analysis. At a five-year follow-up, only tumor size (p = 0.021) and TLI (p = 0.016) individually proved to be significant prognostic indicators of relapse-free survival. Conversely, p27 expression was not related to RFS and c-erbB-2 expression appeared to have only a short-term effect on patient prognosis. TLI and tumor size, tested in multivariate analysis along with ER and menopausal status, maintained their independent prognostic relevance. The study, performed on a large series of node-negative patients given loco-regional treatment alone, for the first time prospectively recruited, showed the prognostic relevance of TLI and its independence from other clinico-pathologic and biologic factors over a five-year period.