Prognostic value of Ki67 in localized prostate carcinoma: a multi-institutional study of >1000 prostatectomies

BACKGROUND

Expanding interest in and use of active surveillance for early state prostate cancer (PC) has increased need for prognostic biomarkers. Using a multi-institutional tissue microarray resource including over 1000 radical prostatectomy samples, we sought to correlate Ki67 expression captured by an automated image analysis system with clinicopathological features and validate its utility as a clinical grade test in predicting cancer-specific outcomes.

METHODS

After immunostaining, the Ki67 proliferation index (PI) of tumor areas of each core (three cancer cores/case) was analyzed using a nuclear quantification algorithm (Aperio). We assessed whether Ki67 PI was associated with clinicopathological factors and recurrence-free survival (RFS) including biochemical recurrence, metastasis or PC death (7-year median follow-up).

RESULTS

In 1004 PCs (∼4000 tissue cores) Ki67 PI showed significantly higher inter-tumor (0.68) than intra-tumor variation (0.39). Ki67 PI was associated with stage (P<0.0001), seminal vesicle invasion (SVI, P=0.02), extracapsular extension (ECE, P<0.0001) and Gleason score (GS, P<0.0001). Ki67 PI as a continuous variable significantly correlated with recurrence-free, overall and disease-specific survival by multivariable Cox proportional hazard model (hazards ratio (HR)=1.04-1.1, P=0.02-0.0008). High Ki67 score (defined as ⩾5%) was significantly associated with worse RFS (HR=1.47, P=0.0007) and worse overall survival (HR=2.03, P=0.03).

CONCLUSIONS

In localized PC treated by radical prostatectomy, higher Ki67 PI assessed using a clinical grade automated algorithm is strongly associated with a higher GS, stage, SVI and ECE and greater probability of recurrence.

Downstream signalling and specific inhibition of c-MET/HGF pathway in small cell lung cancer: implications for tumour invasion

The c-MET receptor can be overexpressed, amplified, or mutated in solid tumours including small cell lung cancer (SCLC). In c-MET-overexpressing SCLC cell line NCI-H69, hepatocyte growth factor (HGF) dramatically induced c-MET phosphorylation at phosphoepitopes pY1230/1234/1235 (catalytic tyrosine kinase), pY1003 (juxtamembrane), and also of paxillin at pY31 (CRKL-binding site). We utilised a global proteomics phosphoantibody array approach to identify further c-MET/HGF signal transduction intermediates in SCLC. Strong HGF induction of specific phosphorylation sites in phosphoproteins involved in c-MET/HGF signal transduction was detected, namely adducin-alpha [S724], adducin-gamma [S662], CREB [S133], ERK1 [T185/Y187], ERK1/2 [T202/Y204], ERK2 [T185/Y187], MAPKK (MEK) 1/2 [S221/S225], MAPKK (MEK) 3/6 [S189/S207], RB [S612], RB1 [S780], JNK [T183/Y185], STAT3 [S727], focal adhesion kinase (FAK) [Y576/S722/S910], p38alpha-MAPK [T180/Y182], and AKT1[S473] and [T308]. Conversely, inhibition of phosphorylation by HGF in protein kinase C (PKC), protein kinase R (PKR), and also CDK1 was identified. Phosphoantibody-based immunohistochemical analysis of SCLC tumour tissue and microarray established the role of c-MET in SCLC biology. This supports a role of c-MET activation in tumour invasive front in the tumour progression and invasion involving FAK and AKT downstream. The c-MET serves as an attractive therapeutic target in SCLC, as shown through small interfering RNA (siRNA) and selective prototype c-MET inhibitor SU11274, inhibiting the phosphorylation of c-MET itself and its downstream molecules such as AKT, S6 kinase, and ERK1/2. Investigation of mechanisms of invasion and, ultimately, metastasis in SCLC would be very useful with these signal transduction molecules.

Alpha-methylacyl-CoA racemase: a multi-institutional study of a new prostate cancer marker

AIM

To test whether alpha-methylacyl-CoA racemase (AMACR) is a sensitive and specific marker of prostate cancer.

METHODS AND RESULTS

The expression levels of AMACR mRNA were measured by real-time polymerase chain reaction. A total of 807 prostatic specimens were further examined by immunohistochemistry specific for AMACR. Quantitative immunostaining analyses were carried out by using the ChromaVision Automated Cellular Imaging System and the Ariol SL-50 Imaging System, respectively. AMACR mRNA levels measured in prostatic adenocarcinoma were 55 times higher than those in benign prostate tissue. Of 454 cases of prostatic adenocarcinoma, 441 were positive for AMACR, while 254 of 277 cases of benign prostate were negative for AMACR. The sensitivity and specificity of AMACR immunodetection of prostatic adenocarcinomas were 97% and 92%, respectively. Both positive and negative predictive values were 95%. By automatic imaging analyses, the AMACR immunostaining intensity and percentage in prostatic adenocarcinomas were also significantly higher than those in benign prostatic tissue (105.9 versus 16.1 for intensity, 45.7% versus 0.02% and 35.03% versus 4.64% for percentage, respectively).

CONCLUSIONS

We have demonstrated the promising features of AMACR as a biomarker for prostate cancer in this large series and the potential to develop automated quantitative diagnostic tests.

Gastrin-releasing peptide is a mitogen and a morphogen in murine colon cancer

Little is known about the factors involved in regulating the appearance, or differentiation, of solid tumors including those arising from the colon. We herein demonstrate that the mitogen gastrin-releasing peptide (GRP) is a morphogen, critically important in regulating the differentiation of murine colon cancer. Although epithelial cells lining the mouse colon do not normally express GRP and its receptor (GRP-R), both are aberrantly expressed by all better differentiated cancers in wild-type C57BL/6J mice treated with the carcinogen azoxymethane. Whereas small tumors in both wild-type and GRP-R-deficient (i.e., GRP-R-/-) mice are histologically similar, larger tumors become better differentiated in the former but degenerate into more poorly differentiated mucinous adenocarcinomas in the latter. This alteration in phenotype is attributable to GRP increasing focal adhesion kinase expression in GRP-R-expressing tumors. Consistent with GRP acting as a mitogen, GRP/GRP-R coexpressing tumors in wild-type animals also contain more proliferating cells than those occurring in GRP-R-/- mice. Yet tumors are similarly sized in animals of either genotype receiving azoxymethane for identical times, a finding attributable to the significantly higher number of apoptotic cells detected in GRP/GRP-R coexpressing cancers. Thus, these findings indicate that although GRP is a mitogen, aberrant expression does not result in increased tumor growth. Rather, the mitogenic properties of GRP are subordinate to it acting as a morphogen, where it and its receptor are critically involved in regulating colon cancer histological progression by promoting a well-differentiated phenotype.

Proliferative activity of oxyphilic (Hurthle) cells in reactive and neoplastic thyroid lersions

The proliferative potential of oxyphilic (Hurthle) cells (HCs) present in neoplastic and non-neoplastic thyroid lesions is uncertain. To estimate the HCs ability to proliferate and to determine whether their proliferative activity correlates with the biologic behavior of different thyroid oxyphilic lesions, we selected 31 cases of chronic lymphocytic (Hashimoto's) thyroiditis and 28 oxyphilic (Hurthle cell) thyroid tumors, including 12 adenomas and 16 carcinomas. Seven histologically normal thyroid specimens from euthyroid patients served as control tissue. The proliferative activity of HCs was evaluated by means of a double immunostaining for Ki67 and a mitochondrial antigen (which specifically recognizes oxyphilic cells). Oxyphilic cells in thyroiditis had a low proliferative activity (PA: 0.55%), although higher than that of normal thyroid parenchyma (PA: 0.06%). Neoplastic HC lesions had a mean proliferative activity of 1.56% and 6.26% in adenomas and carcinomas, respectively. A statistically significant difference was observed between proliferative activity of non-neoplastic and neoplastic lesions (p<0.01), but not within the tumor group, between adenomas and carcinomas. In addition, HC carcinomas had a statistically significant positive correlation between proliferative activity and tumor size (p<0.01) and the presence of necrosis (p<0.001).