Molecular Markers PCNA, Ki-67 and Cathepsin D in Patients with Colorectal Cancer and their Relationship with Lymph Node Involvement

Colorectal cancer has emerged as a major public health concern globally. It represents a heterogeneous disease resulting from the occurrence of several oncogenic events. Its growth and aggressiveness, and subsequently the likelihood of lymph node involvement, mainly depends on both tumor cell apoptosis and proliferative activity, revealed by several molecular markers. The aim of this study was to assess the usefulness of (1) proliferating cell nuclear antigen (PCNA), a nonhistamine nuclear protein, (2) Ki-67 antigen, and (3) cathepsin D, a lysosomal hydrolase involved in intra- and extracellular proteolysis, in differentiating between node positive (N1) and node negative (N0) patients. PCNA is a specific marker of cell division, Ki-67 is a marker of mitotic index, while cathepsin D is an enzyme aberrantly produced and processed in malignancy, and its hyper-secretion may lead to tumor progression and metastasis. Methods: Data from a group of 58 consecutive patients who underwent surgery for colorectal cancer were reviewed, and pathological specimens re-evaluated. There were 36 (62.1%) men and 22 (37.9%) women, with a median age of 72 years (range 41-76 years). Patients with history of previous cancer were excluded from the study, as well as those who have undergone preoperative adjuvant chemotherapy. Ethical permission and written informed patient consent were obtained. Consecutive 4 lm-thick sections were cut from paraffin-embedded blocks, and immunolabeled. The primary antibodies were (1) anti-Ki-67 (monoclonal anti-human Ki-67 antigen/clone MIB-1), (2) anticathepsin D monoclonal antibody, and (3) anti-PCNA (monoclonal mouse antiproliferating cell antigen/clone PC10). Ki-67 labeling index (Ki-67 LI), cathepsin D expression, and PCNA positivity were evaluated in each specimen, and TNM staging, histologic grade, and vessel involvement were also recorded. Both PCNA and Ki-67 LI were considered positive when the reaction was present in >50% of cells, while cathepsin D expression was considered positive when at least 15% of stromal cells were stained. The X2 test and the Pearson’s correlation coefficient calculation were used, and the p-value <0.01 was considered significant. Results: The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy in predicting lymph node metastases were 63.3%, 89.2%, 77.8%, 80.5%, and 80.0% (OR= 14.44, 95% CI 3-73-55.87, p<0.01) for PCNA; 77.3%, 89.2%, 80.9%, 86.4%, and 85.0% (OR=28.05, 95% CI 6.65-118.28) for Ki-676; and 68.2%, 91.9%, 83.3%, 82.9%, and 83.0% (OR=24-20, 95% CI 5.51-116.97) for cathepsin D, respectively. A strong correlation (R=0.81, p=0.003) between PCNA and Ki-67 was found. The combination of three markers reached 79.3% sensitivity and 100% specificity. Multivariate analysis using a logistic regression model showed that only histologic grade, PCNA, and cathepsin D positivity independently correlated with node metastases. Conclusions: Molecular markers PCNA, Ki-67 and cathepsin D in combination are useful in predicting the risk of lymph node metastases in patients with colorectal cancer, and should be considered reliable risk factors in prediction curability of the disease.