PROBING BREAST CANCER THERAPEUTIC RESPONSES BY DNA CONTENT PROFILING

Background. Discrepancies in the interpretation of breast cancer therapeutic responses still exist mainly because of lack of standardized assessment criteria and methods. Objective. DNA content profiling of cells in the affected (cancerous) tissue before and after neoadjuvant chemotherapy (NAC) was applied to facilitate interpretation of therapeutic responses. Methods. Both diagnostic biopsy and operation materials representing the tissue of primary tumors surgically removed after NAC were subjected to DNA image cytometry. Polyploidy and aneuploidy in DNA histograms were evaluated with a prognostic Auer typing. Stemline DNA index (DI) values and percentages of cells that polyploidize (>4.5C) were also determined. Immunofluorescence staining was applied to evaluate proliferation (Ki-67), invasiveness (CD44), and self-renewal factors characteristic for stem cells (SOX2 and NANOG). Results. DNA content profiles of 12 breast cancer cases, of which 7 were triple-negative, revealed the features of tumor non-responsiveness to NAC in 7 cases, of which 5 were triple-negative. Among non-responsive cases there were 3 cases that showed enhanced polyploidization, suggesting the negative NAC effect. Near-triploid (DI=1.26-1.74) triple-negative cases were determined as most resistant to NAC. Cycling near-triploid cells may contribute to the excessive numbers of >4.5C cells. Polyploid cells were positive for Ki-67, CD44, SOX2, and NANOG. Conclusions. DNA content profiling data provide additional helpful information for interpreting therapeutic responses in NAC-treated breast cancers. Polyploid tumor cells possessing stem cell features can be induced by NAC. Because NAC effects in some cases may be unfavorable, the use of the further treatment strategy should be carefully considered.

Abstract P3-03-17: Can polyploid tumor cells possessing stem cell features be induced in resistant breast carcinomas?

Cancer stem cells are believed to be responsible for radio- and chemoresistance of malignant tumors. In vitro studies demonstrate that ionizing radiation is capable of reprogramming cancer cells from non-stem state into stem state [1, 2]. Moreover, the embryonic stemness cassette was found to be expressed in tumor cells (including breast cancer cells) after they were polyploidized as a result of genotoxic stress [2, 3], thus prompting us to suppose that the polyploid cells and their descendants released by depolyploidization can possess stem cell characteristics. The aim of the current work is to test whether polyploid cells having stem cell features can be also induced in vivo, namely in locally advanced breast carcinomas as a result of neoadjuvant chemotherapy (NAC), assuming that this process is not autonomous, but rather stipulated by the tumor microenvironment.

The study population consisted of 30 breast cancer patients of age ranged from 31 to 75 y. o. diagnosed in the Latvian Oncology Center of the Riga East University Hospital between 2013 and 2015. The tissue specimens were collected after the patients' informed consent was obtained in accordance with the Ethics regulations. The clinico-pathologic information about these patients, including Ki-67 index and the status of ER, PR and HER2 receptors, was obtained from the aforementioned clinics. The majority of patients (n = 28) had locally advanced breast cancer, predominantly Stage III disease. Both diagnostic biopsy and operation material, such as primary tumors surgically removed after NAC using standard doses of paclitaxel and doxorubicin, were subjected to DNA content analysis with image cytometry. Ploidy-related parameters, such as DNA index and the percentage of cells exceeding 4.5c (presumably proliferating and polyploid cells), were determined. Immunofluorescence staining was applied to evaluate expression of such markers/factors as proliferation (Ki-67), stemness (SOX2 and NANOG) and invasiveness (CD44).

At the time of diagnosis, 14 patients had primary tumors possessing near-triploid clones, and these cases in comparison with 16 other cases comprised of near-euploid clones had 4.5-fold increase of percentages of cells exceeding the ploidy of 4.5c (p < 0.05) and 1.3-fold increase of percentages of cells positive for Ki-67 (p > 0.05). Of 10 cases diagnosed as “triple-negative”, 6 were near-triploid. Among those cases that showed the resistance to NAC (grades 1 and 2 by Miller-Payne histopathologic scoring), 67% were near-triploid. Polyploidization, which in some resistant cases is gained by NAC, was likely to be attributed to near-triploid clones. Notably, polyploid cells were positive for Ki-67, SOX2, NANOG, and CD44. Thus, these non-quiescent polyploid cells can possess the invasiveness and self-renewal features that were also seen in descendants after depolyploidization. Perhaps, reversible polyploidy plays a definite role in gaining the resistance of tumor cells to chemo- and radiotherapy in vivo.

1. Ghisolfi L. et al. PLoS ONE 2012; 7: e43628.

2. Lagadec C. et al. Stem Cells 2012; 30: 833-44.

3. Salmina K. et al. Exp Cell Res 2010; 316: 2099-112.

Citation Format: Gerashchenko BI, Salmina K, Eglitis J, Erenpreisa J. Can polyploid tumor cells possessing stem cell features be induced in resistant breast carcinomas? [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P3-03-17.

Disentangling the aneuploidy and senescence paradoxes: a study of triploid breast cancers non-responsive to neoadjuvant therapy

Aneuploid cells should have a reduced proliferation rate due to difficulty in proceeding through mitosis. However, contrary to this, high aneuploidy is associated with aggressive tumour growth and poor survival prognosis, in particular in triploid breast cancer. A further paradox revolves around the observation that, while cell senescence should inhibit proliferation, the senescence marker p16INK4a correlates with poor treatment outcome in patients with a very aggressive triple-negative breast carcinoma (TNBC). In this study, we aim to pour light on the possible relationship of these conundrums with polyploidy of tumour cells. We performed detailed analysis of DNA histogram profiles in diagnostic core biopsies of 30 cases of operable breast cancer and found that near triploidy in TNBC and other forms correlated with weak or no response to neoadjuvant chemotherapy (NAC) as scored by Miller-Payne index. Polyploid cells in operation samples from tumours that were non-responsive to NAC treatment were Ki67 and CD44 positive. In addition, polyploid cells were positive for markers of embryonic stemness (OCT4, SOX2, NANOG) and senescence (p16INK4a). The relationship patterns between p16INK4a and NANOG were heterogeneous, with predominantly mutually exclusive expression but also synergistic and intermediate variants in the same samples. We conclude that the aneuploidy and senescence paradoxes can be explained by the mutual platform of polyploidy, conferring genomic and epigenetic instability as a survival advantage. Such cells are able to bypass aneuploidy restrictions of conventional mitosis and overcome the barrier of senescence by a shift to self-renewal, resulting in progression of cancer.

Trefoil factor 3 is required for differentiation of thyroid follicular cells and acts as a context-dependent tumor suppressor

Trefoil factor 3 (TFF3) is overexpressed in a variety of solid epithelial cancers, where it has been shown to promote migration, invasion, proliferation, survival and angiogenesis. On the contrary, in the majority of thyroid tumors, it is downregulated, yet its role in the development of thyroid cancer remains unknown. Here we show that TFF3 exhibits strong cytoplasmic staining of normal thyroid follicular cells and colloid and the staining is increased in hyperfunctioning thyroid nodules, while it is decreased in all thyroid cancers of follicular cell origin. By meta-analysis of gene expression datasets, we found that in the thyroid cancer, conversely to the breast cancer, the expression of TFF3 mRNA was downregulated by estrogen signaling and confirmed this by treating thyroid cancer cells with estradiol. Forced expression of TFF3 in anaplastic thyroid cancer cells resulted in decreased cell proliferation, clonal spheroid formation and entry into the S phase. Furthermore, it induced acquisition of epithelial-like cell morphology and expression of the differentiation markers of thyroid follicular cells and transcription factors implicated in the thyroid morphogenesis and function. Taken together, this study provides the first evidence that TFF3 may act as a tumor suppressor or an oncogene depending on the cellular context.

Abstract P5-02-02: Factors influencing time to seeking medical advice and start of treatment in breast cancer (BC) patients – an International survey

Background. We earlier reported patient-related factors influencing the delay in first medical advice for signs of BC in selected countries (J Clin Oncol 2012;30[15S]:578s). The present analysis, which includes additional countries, addresses the factors influencing time from first symptoms of BC to initiation of treatment (Total Delay Time; TDT).

Methods. A total of 6,588 female BC patients from 11 countries were surveyed using a uniform questionnaire translated into local languages. TDT was determined using 8 individual scales, including one pertaining to patient delay and 7 related to subsequent steps in a typical diagnostic process. Regression models were constructed using 18 variables concerning diverse contextual and personal patient characteristics. Due to expected differences in the relevant sets of predictors, time between first symptoms and first medical visit (Patient Delay Time; PDT) and time between first medical visit and start of therapy (System Delay Time; SDT) were modeled separately, with multilevel regression.

Results. Mean TDT varied in individual countries from 11.5 to 29.4 weeks (grand mean of 14.3 weeks; see table), with 43% of cases with a delay of >12 weeks. Multilevel regression equation indicated that factors significantly correlated with longer PDT were distrust in the medical system and ignoring disease. Patients with fear of the disease, stronger self-examination habits, at least secondary education, being employed, reporting more support from friends and family, and living in cities of >100,000 inhabitants had shorter PDT. Predictors of shorter SDT included being diagnosed by an oncologist vs. other physician, having at least secondary education, being older than 60 years, having a history of cancer in female relatives and having breast lump vs. other BC symptoms. Indicators of longer SDT were PDT >4 weeks, more than one BC symptom detected by patient, distrust in the medical system, disregard of BC signs, less support from friends and family, and having first medical examination in a public vs. private medical center. Individual countries differed significantly with regard to intercept of the multilevel models and slopes of regression coefficients for selected psychological and behavioral attributes.

Conclusions. An extensive set of variables potentially impacting delay times, mainly related to psychological and behavioral patient attributes, was examined. Several of them strongly determined both PDT and SDT, but their strength differed between individual countries. Both models (for PDT and SDT), although statistically significant, accounted for approximately 20% of variance in time; therefore other variables, e.g. related to the differences in national healthcare systems (not addressed in this study) might have a stronger impact on delays in the initiation of BC therapy and warrant further analyses.