**Background:** Colorectal cancer (CRC) remains a major public health problem worldwide. Although TNM staging is the most important prognostic factor, it provides limited information, as some early-stage patients have aggressive disease while some advanced-stage patients have favorable outcomes. There is a need for additional, easily assessable histopathological parameters that can predict lymph node metastases and aggressive behavior. This study aimed to evaluate the prognostic significance of classic factors and less-studied parameters—tumor budding (TB), poorly differentiated clusters (PDCs), tumor-infiltrating lymphocytes (TILs), and tumor border configuration—on routine hematoxylin–eosin (HE) slides.
**Methods:** A retrospective study was conducted on 71 CRC patients who underwent surgery at the “Pius Brînzeu” County Clinical Emergency Hospital in Timișoara, Romania. The cohort included 50 consecutive cases from 2014 and 21 robotically operated cases from July 2015 to July 2016. Exclusion criteria were diagnoses on biopsies, non-carcinoma tumors, neoadjuvant therapy, and tumor recurrences. Clinical and morphological data were collected from pathology records. HE-stained slides were re-evaluated by two junior pathologists (double-blind) with discordant results resolved by a senior pathologist. Parameters analyzed included: age, sex, tumor location (right colon, left colon, rectum), histological type, WHO grade (based on gland formation: G1–G4, with binary low-grade [G1–G2] vs. high-grade [G3–G4]), TB grade (GBd: G1Bd 0–4 buds, G2Bd 5–9, G3Bd ≥10 per 0.785 mm² field at 200×), PDCs grade (PDCs-G: G1 <5, G2 5–9, G3 ≥10 clusters), depth of invasion (pT1–pT2 vs. pT3–pT4), lymph node status (pN0 vs. pN+), distant metastases (pM0 vs. pM1), AJCC stage, lymphovascular invasion (LVI), perineural invasion (PNI), tumor necrosis (<10% vs. ≥10%), ulceration, TILs (TIL− ≤5% vs. TILs+ >5% mononuclear cells at 400×), and tumor border configuration (pushing vs. infiltrative). Statistical analysis used Pearson correlation coefficient (r) with p<0.05 considered significant. Multivariate analysis was performed only for the 69 conventional adenocarcinoma (ADK NOS) cases, excluding 2 mucinous adenocarcinomas.
**Key Results:** The cohort comprised 44 men (62.0%) and 27 women (38.0%), mean age 66.47 years; 60.6% were ≥65 years. Tumors were located in the right colon (21.1%), left colon (32.4%), and rectum (46.5%). Most were conventional ADK NOS (69 cases), with 2 mucinous ADKs. Deep invasion (pT3–pT4) was present in 84.5% of cases. Lymph node metastases (pN+) were found in 47.9% (34 cases), distant metastases in 8.4% (6 cases), LVI+ in 39.4% (28 cases), and PNI+ in 32.4% (23 cases). Infiltrative tumor border was seen in 63.4% (45 cases), and TILs+ in 84.5% (60 cases).
GRADING DISTRIBUTION
WHO grade—G1 8.7%, G2 79.7%, G3–G4 11.6%; GBd—G1Bd 23.2%, G2Bd 27.5%, G3Bd 49.3%; PDCs-G—G1 16%, G2 44.9%, G3 39.1%.
Multivariate analysis showed significant positive correlations (all p<0.0001 unless noted) with lymph node metastases (pN+) for: high WHO grade (p=0.006), deep invasion pT3–pT4 (p=0.003), advanced AJCC stage (p<0.0001), LVI (p<0.0001), PNI (p<0.0001), high-grade TB (p<0.0001), high-grade PDCs (p<0.0001), and infiltrative tumor border (p<0.0001). GBd also correlated with pM (p=0.019) and tumor necrosis (p=0.009). PDCs-G correlated with pT (p<0.0001). Infiltrative border correlated with right colon location (p=0.036), high WHO grade (p=0.001), high GBd (p<0.0001), high PDCs-G (p<0.0001), pT3–pT4 (p<0.0001), pN+ (p<0.0001), pM1 (p=0.047), advanced stage (p<0.0001), LVI+ (p<0.0001), PNI+ (p<0.0001), and tumor necrosis (p=0.040). TILs did not show significant correlations with any other parameter.
**Clinical Implications:** This study demonstrates that easily assessable histopathological parameters on routine HE slides—particularly high-grade tumor budding, high-grade poorly differentiated clusters, infiltrative tumor border configuration, and perineural invasion—are strongly associated with lymph node metastases and other adverse prognostic factors in colorectal cancer. These markers could complement traditional TNM staging to better identify patients at high risk for aggressive disease, potentially guiding more intensive treatment strategies. The findings support the inclusion of TB, PDCs, and tumor border configuration in routine pathology reporting, as recommended by the International Tumor Budding Consensus Conference (ITBCC) 2016. However, the lack of significant correlation for TILs in this study highlights the need for standardized assessment methods. Limitations include the retrospective single-center design, small sample size (n=71), and absence of survival data. Larger prospective studies with clinical follow-up are needed to validate these parameters as independent prognostic factors and to establish standardized quantification schemes.