The specificity and sensitivity of pro-gastrin-releasing peptide (ProGRP) was evaluated in 37 healthy subjects, 197 patients with benign diseases and 310 patients with malignant diseases of different origins. Abnormal ProGRP serum levels (>50 pg/ml) were found in 10% of the patients with benign diseases and in 26.1% of the patients with active cancer. None of the healthy subjects had abnormal ProGRP levels. The benign disease with the highest ProGRP concentration was renal failure, with abnormal values in 51.6% of the patients studied. Excluding patients with renal failure or patients with creatinine levels greater than 1.5 mg/dl, raised ProGRP values (<80 ng/ml) were found in 2.5% (4/160) of patients with benign diseases and in 4.9% of patients with active malignancies other than lung cancer or neuroendocrine tumors (<110 ng/ml). Abnormal ProGRP serum levels were found in 26.2% of patients with non-small cell lung cancer (NSCLC) (mean 40.5 ± 35.4 pg/ml) and in 76.8% of patients with small cell lung cancer (SCLC) (mean 694 ± 1,776 pg/ml) (p < 0.001). ProGRP serum levels >300 pg/ml were only found in SCLC patients (41.4%). ProGRP results were related to tumor extension in SCLC (sensitivity in limited disease 58.3%, in extensive disease 95.5%) but not in NSCLC. In summary, renal failure is the most frequent source of false-positive results with ProGRP, and this marker is useful in the histological differential diagnosis of lung cancer.

Carcinoembryonic antigen (CEA) and squamous cell carcinoma(SCC) serum levels were prospectively determined in 159 untreated patients diagnosed with carcinoma of the uterine cervix from 1991 to 2001. The histological analysis showed epidermoid cancer in 117 patients, adenocarcinoma in 26 patients, adenosquamous carcinoma in 12 patients and other histological types in the remaining 4 patients. Tumor marker sensitivity was related to the histological type with abnormal SCC (>2 ng/ml) in 51.3% of squamous tumors in contrast to the 7.1% found in other histologies. By contrast, CEA sensitivity was not related to histology with abnormal values (>5 ng/ml) in 25% of squamous tumors, 19% of adenocarcinomas, 33% of adenosquamous carcinomas and 25% of other histologies. CEA and SCC serum levels were clearly related to tumor stage, parametrial invasion, tumor size and nodal involvement. Elevated pretreatment CEA indicates parametrial invasion with a probability of 82%. Likewise, pretreatment CEA and SCC serum levels were of prognostic value, with a shorter disease-free survival and overall survival in patients with abnormal levels. All patients with adenocarcinomas and abnormal CEA had relapse during follow-up. Multivariate analysis indicated that parametrial invasion, age, tumor size and SCC were independent prognostic factors. In conclusion, CEA and SCC are useful tumor markers in carcinomas of the uterine cervix, with a clear relationship with well-known prognostic factors (parametrial invasion, nodal involvement), and are of prognostic value.