Effects of different preprocessing algorithms on the prognostic value of breast tumour microscopic images.

The purpose of this study was to improve the prognostic value of tumour histopathology image analysis methodology by image preprocessing. Key image qualities were modified including contrast, sharpness and brightness. The texture information was subsequently extracted from images of haematoxylin/eosin-stained tumour tissue sections by GLCM, monofractal and multifractal algorithms without any analytical limitation to predefined structures. Images were derived from patient groups with invasive breast carcinoma (BC, 93 patients) and inflammatory breast carcinoma (IBC, 51 patients). The prognostic performance was indeed significantly enhanced by preprocessing with the average AUCs of individual texture features improving from 0.68 ± 0.05 for original to 0.78 ± 0.01 for preprocessed images in the BC group and 0.75 ± 0.01 to 0.80 ± 0.02 in the IBC group. Image preprocessing also improved the prognostic independence of texture features as indicated by multivariate analysis. Surprisingly, the tonal histogram compression by the nonnormalisation preprocessing has prognostically outperformed the tested contrast normalisation algorithms. Generally, features without prognostic value showed higher susceptibility to prognostic enhancement by preprocessing whereas IDM texture feature was exceptionally susceptible. The obtained results are suggestive of the existence of distinct texture prognostic clues in the two examined types of breast cancer. The obtained enhancement of prognostic performance is essential for the anticipated clinical use of this method as a simple and cost-effective prognosticator of cancer outcome.

Prognostic factors for longer disease free survival and overall survival after surgical resection of isolated liver metastasis from breast cancer.

PURPOSE: Isolated liver metastases (LMs) from breast cancer (BC) occur in only 1-3% of the cases. Resection of isolated LMs improves survival. We examined the prognostic factors for time to LM development, disease free survival (DFS) and overall survival (OS) after BCLM resection. METHODS: From 2006 to 2009, 32 patients underwent LM resection. All of them had breast cancer surgery for their primary tumor and developed resectable LMs as the first and only site of disease progression. RESULTS: LMs developed after a median of 25 months. With a median follow up of 37 months (range 7-66) after metastases resection, median DFS and OS (with 95% CI) were 22.5 (12-40) and 37 (≥23) months, respectively. Tumor size ≥3 vs <3 cm and adjuvant chemotherapy vs no adjuvant chemotherapy correlated with shorter time to LM development (p<0.01 for both parameters). These parameters and BC negative estrogen (ER)/ progesterone receptors (PR) (ER?/PR? vs other) were related with shorter DFS. Positive (vs negative) axillary lymph nodes and BC negative ER/ PR (ER?/PR? vs other) status correlated with shorter OS (p<0.01 for both parameters). A period to metastases development ≥ 24 months (vs ≤24) and single (vs multiple) metastases were related with longer DFS and OS (p<0.01 for both conditions). CONCLUSION: Despite the relatively small number of patients in this study, we believe that positive ER/PR status for both BC and LMs, negative axillary lymph nodes, time to liver metastases development >24 months and single liver metastases predict longer DFS and OS after LM resection.

Clinical and pathological response to induction chemotherapy used as a prognostic factor in inflammatory breast cancer. Single institution experience.

PURPOSE: To evaluate clinical and pathological characteristics of patients with inflammatory breast carcinoma (IBC). Also, to evaluate the importance of achieved clinical and pathological responses to induction chemotherapy (iCT) and their role in the prognosis of IBC. METHODS: The medical records of 81 female patients with stage IIIB IBC, diagnosed between January 2008 and December 2010 at the Institute for Oncology and Radiology of Serbia (IORS) were evaluated. Almost all of the patients received anthracycline-based iCT. After 3-4 cycles of iCT, the clinical response (defined as complete response/CR, partial response/PR, stable disease/SD and disease progression/ PD) was assessed. Also, pathological response to iCT (defined as pathological complete response/pCR, near complete response/pNCR, partial response/pPR and no change/ pNC) was estimated in patients who had undergone surgery. All first metastatic sites were recorded. RESULTS: Clinical CR/PR was observed in 61.8% of the patients, while the pathological response (pCR, pNCR/near complete response, and pPR) rate in patients who had undergone surgery was 70%. During follow-up 22 (27.2%) patients developed PD (8 responders and 14 non-responders). Most common metastatic sites were the skeleton in non-responders and the liver in responders. Central nervous system (CNS) metastases developed in 24% of non-responders while no responder developed such metastases. Non-responders had shorter OS compared to responders, but without statistical significance. CONCLUSION: Although the number of the patients analysed in this study is relatively small, we believe that response to iCT could be used as a prognostic marker, since patients who initially failed to respond to iCT showed a higher risk for PD with development of distant metastases, primarily in bones and CNS, and shorter survival.

Incidence of brain metastases in early stage HER2 3+ breast cancer patients; is there a role for brain CT in asymptomatic patients?

PURPOSE: Human epidermal growth factor receptor 2 overexpression (HER2 3+) is reported in retrospective studies as a factor that contributes to higher incidence of brain metastases (BM) in patients with metastatic breast carcinoma. Although there are some reports suggesting higher incidence of BM in adjuvant trastuzumab trials, the true incidence, as well as the time of occurrence of BM in early-stage high risk breast carcinoma patients, has not been widely prospectively explored. The main objective of this study was to prospectively explore the incidence of BM during and after adjuvant trastuzumab administration in HER2 3+ early-breast carcinoma patients. METHODS: Two hundred and fifty-eight patients with early, HER2 3+ breast carcinoma have been included in this analysis. Brain computed tomography (CT) scan was scheduled once during adjuvant trastuzumab therapy and thereafter only if central nervous system (CNS) symptoms occurred. RESULTS: Eighty-five patients (33%) underwent brain CT in the absence of CNS symptoms. The median number of trastuzumab cycles at the time of brain CT was 9 (range 4-18). There were no BM detected by brain CT in these 85 asymptomatic patients. However, during a median follow up of 18 months 5/258 patients (1.93%) developed BM, but only 2 (0.77%) while still receiving adjuvant trastuzumab. The median time from breast cancer diagnosis to BM was 24 months (range 14-43). CONCLUSION: BM are a rare event during adjuvant trastuzumab treatment and brain CT screening is not justified in asymptomatic patients with early HER2 3+ breast carcinoma.

Do we know how many cancer patients have a family history of cancer?

PURPOSE: It has been estimated that approximately 5-10% of the general population have a family history that is indicative of hereditary cancer, predominately breast and colorectal. However, it is not precisely known how many patients have positive family history of cancer. The purpose of this study was to determine how many cancer patients have positive family history of cancer. METHODS: Patients were interviewed during the first visit to Daily Chemotherapy Hospital (DCH) of the Institute for Oncology and Radiology of Serbia, Belgrade. Data about patient cancer type and cancer types among family members were recorded in the hospital chart and analyzed. RESULTS: During an 8-month period, 677 newly diagnosed cancer patients with 9 cancer types were referred to DCH for chemotherapy. Positive family history (at least one first degree relative) for any cancer type was recorded in 163 (24.1%) patients and in 47 (6.9%) patients for the same cancer type. The highest percentage of the positive family history for the same type of cancer showed patients with breast cancer (9.9%), followed by colorectal (7.2%) and brain tumors (6.25%). CONCLUSION: The overall incidence of positive family cancer history was 31.0% and was higher than expected. Cancer can be more disturbing for persons who already had experience with this disease in a close family member. Those patients need special attention with more intensive and carefully preplanned psychological support.