Examination of tumor biological factors for prognostic and predictive indicators is not part of routine testing in ovarian cancer. As in other tumors, the detection of hematogenous tumor spread could help to estimate the risk of metastatic disease. We examined the expression of p53, KI67, topoisomerase IIα (Top IIa), epidermal growth factor receptor (EGFR), human epithelial growth factor receptor 2 (HER2) and nm23 in tumor tissues from 90 patients with ovarian cancer. All underwent bone marrow (BM) aspiration and screening for disseminated tumor cells in the bone marrow (DTC-BM) at primary diagnosis. BM aspiration, cytospin preparation, and immunocytochemical staining with the anticytokeratin antibody (A45-B/B3) were done following a standardized protocol. The expression of p53, KI67, Top IIa, EGFR, HER2, and nm23 was evaluated by immunohistochemistry on paraffin-embedded tissue samples and classified by percentage of stained cells or immunoreactive score (IRS). The prognostic impact of the individual factors together with standard histologic parameters was calculated by univariate and multivariate analyses. Expression rates for HER2 (2+/3+: 34.5%), KI67 (median 30%), p53 (median IRS 5), and Top IIa (median IRS 4) were relatively high, whereas nm23 (median IRS 2) and EGFR (IRS 0: 61%) showed weak staining. In 21/90 patients (23.3%), DTC-BM (≥1/2 × 106 cells) could be detected. The presence of DTC-BM was inversely related to nodal status (P= .015) but not to the other factors examined. Tumor stage (P= .02), lymph node involvement (P= .003), grade (P= .046), postoperative tumor residue (P <.001), peritoneal seeding (P= .02), and KI67 (P= .046) significantly correlated with overall survival (OS) after a median observation time of 28 months (2–105). The finding of ascites was borderline significant (P= .050). The presence of DTC-BM (P= .04) and KI67 positivity (P= .02) predicted reduced distant disease-free survival. By multivariate analysis, postoperative tumor residue remained an independent factor for OS (P= .02, relative risk = 4.6). As a primarily locoregional disease, tumor stage and postoperative tumor residue are the main determinants of prognosis in patients with ovarian cancer. However, even in advanced stages, examination of tumor biological factors could help to stratify subgroups of patients and establish targeted therapies.

Beim Ovarialkarzinom ist die Bestimmung tumorbiologischer Faktoren zu prognostischen und prädiktiven Zwecken wenig etabliert. Durch eine Detektion hämatogen gestreuter Tumorzellen ließe sich das Risiko einer Fernmetastasierung abschätzen. Wir untersuchten die Expression von p53, KI67, HER2, Topoisomerase IIa, EGFR und nm23 an Ovarialkarzinom-Geweben von 90 Pat, bei denen im Rahmen der Primäroperation eine Untersuchung auf disseminierte Tumorzellen im Knochenmark (DTZ-KM) erfolgt war. Die KM-Aspiration, Zytospinpräparation und immunzytochemische Färbung mit dem anti-Zytokeratin Antikörper A45 B/B3 erfolgte nach einem standardisierten Protokoll. Die Expression von p53, KI67, HER2, Topoisomerase IIa, EGFR und nm23 wurde durch Immunhistochemie an Paraffinschnitten bestimmt, eine Klassifikation erfolgte nach Prozentsatz gefärbter Zellen bzw. immunreaktivem Score (IRS). Die Expressionsraten von HER2 (2+/3+: 34%), KI67 (med. 30%), p53 (med IRS 5) und Topoisomerase IIa (med IRS 4) waren hoch, während nm23 (med IRS 2) und EGFR (med IRS 0) nur schwach exprimiert wurden. Bei 21 Pat (23%) fanden sich DTZ-KM. Deren Präsenz war lediglich mit dem Nodalstatus korreliert (p=0,05). Während das Tumorstadium (p=0,02), Lymphknotenbefall (p=0,003), Grading (p=0,05), postoperativer Tumorrest (p<0,001), Aszites (p=0,05), Peritonealkarzinose (p=0,02) und KI67 (p=0,05) jeweils signifikant mit dem Gesamtüberleben korrelierten, zeigte die Präsenz von DTZ-KM (p=0,04) und KI67- Positivität (p=0,02) ein erhöhtes Risiko einer Metastasierung auf. In der multivariaten Analyse verblieb lediglich der postoperative Tumorrest als unabhängiger Faktor für das Metastasen freie und Gesamtüberleben (p=0,02, RR=4,6). Beim Ovarialkarzinom besitzt weiterhin das Tumorstadium und der postoperative Tumorrest entscheidende Bedeutung für die weitere Prognose. Dennoch ließen sich durch die Bestimmung tumorbiologischer Faktoren Subgruppen definieren und möglicherweise zielgerichtete Therapien etablieren.

Background and objective: In the last few years special attention has been paid to the serum concentration of haemoglobin in patients with cancer. It was the aim of this study to ascertain whether at the time of the initial diagnosis low haemoglobin levels in patients with breast cancer denote a higher risk of primary haematogenous dissemination of tumour cells in bone marrow than that in those with higher levels.Patients and methods: Between March 1994 and March 2000 bone marrow aspirates were performed and serum haemoglobin concentrations (g/dl) measured before primary surgical treatment in 360 consecutive patients (mean age 57.5 years) with primary breast cancer. Evidence of isolated tumour cells in bone marrow was obtained with the pancytokeratin antibody A45-B/B3. The cohort was divided into two groups on the basis of mean haemoglobin values, and the patients underwent follow-up examination a mean of 30.7 months after the initial diagnosis. Patients with metastases at first diagnosis or those who had received nonsurgical treatment at that time were excluded.Results: The mean pre-treatment haemoglobin concentration of the cohort was 13.8 g/dl (median 13.9 g/dl, S.D.1.2). There was no statistically significant difference in the frequency of cytokeratin-positive bone marrow findings between the two groups. While disseminated tumour cells were demonstrated in the bone marrow of 48 (28%) patients with a pre-treatment haemoglobin of > 13.9 g/dl (p = 0.50), this was so in 58 patients (31%) with a pre-treatment haemoglobin of > 13.9 g/dl. There was also no difference between the two groups regarding median survival time (67.9 vs. 65.8 months; p = 0.46). However, there was a significant difference in probability of survival between patients with or without isolated tumour cells in the bone-marrow (59.7 vs. 69.2 months; p < 0.0001).Conclusion: There is no evidence at present that the preoperative haemoglobin concentration is of prognostic value regarding the haematogenous dissemination of tumour cells and the survival time of patients with primary breast cancer.

There is no evidence at present that the preoperative haemoglobin concentration is of prognostic value regarding the haematogenous dissemination of tumour cells and the survival time of patients with primary breast cancer.

Hintergrund und Fragestellung: Patienten und Methodik: Ergebnisse: Folgerung: Background and objective: Patients and methods: Results: Conclusion: