A novel machine learning-derived decision tree including uPA/PAI-1 for breast cancer care.

Background uPA and PAI-1 are breast cancer biomarkers that evaluate the benefit of chemotherapy (CT) for HER2-negative, estrogen receptor-positive, low or intermediate grade patients. Our objectives were to observe clinical routine use of uPA/PAI-1 and to build a new therapeutic decision tree integrating uPA/PAI-1. Methods We observed the concordance between CT indications proposed by a canonical decision tree representative of French practices (not including uPA/PAI-1) and actual CT prescriptions decided by a medical board which included uPA/PAI-1. We used a method of machine learning for the analysis of concordant and non-concordant CT prescriptions to generate a novel scheme for CT indications. Results We observed a concordance rate of 71% between indications proposed by the canonical decision tree and actual prescriptions. Discrepancies were due to CT contraindications, high tumor grade and uPA/PAI-1 level. Altogether, uPA/PAI-1 were a decisive factor for the final decision in 17% of cases by avoiding CT prescription in two-thirds of cases and inducing CT in other cases. Remarkably, we noted that in routine practice, elevated uPA/PAI-1 levels seem not to be considered as a sufficient indication for CT for N≤3, Ki 67≤30% tumors, but are considered in association with at least one additional marker such as Ki 67>14%, vascular invasion and ER-H score <150. Conclusions This study highlights that in the routine clinical practice uPA/PAI-1 are never used as the sole indication for CT. Combined with other routinely used biomarkers, uPA/PAI-1 present an added value to orientate the therapeutic choice.

A prospective study to assess the clinical utility of serum HER2 extracellular domain in breast cancer with HER2 overexpression.

PURPOSE: We explored the clinical utility of human epidermal growth factor receptor-2 extracellular domain (HER2/ECD) in patients treated for an invasive breast cancer with HER2 overexpression. METHODS: We prospectively studied HER2/ECD levels in the sera of 334 women included between 2007 and 2014, all treated with trastuzumab. HER2/ECD levels were measured at diagnosis, during treatments, and along the follow-up. We investigated the relationship of HER2/ECD with other clinicopathological parameters at diagnosis, its prognosis value, and its utility during the monitoring of a neoadjuvant treatment and the follow-up. RESULTS: Elevated HER2/ECD at diagnosis correlated positively with parameters associated with tumor aggressiveness. Disease-free survival of non-metastatic patients was significantly shorter in patients with high HER2/ECD at diagnosis (HR = 13.6, 95 % CI 1.6-113.6, P < 0.0001). Progression-free survival of metastatic patients was better for patients with low HER2/ECD (HR = 2.6, 95 % CI 1.2-5.3, P = 0.033). A multivariate analysis revealed that HER2/ECD level at diagnosis was an independent prognosis factor. During neoadjuvant therapy, a significant decrease in HER2/ECD was reported only for the complete histological response group (P = 0.031). During the follow-up, HER2/ECD helped predict relapse, disease progression, and metastases before imaging in 18.6 % cases of the studied cohort. CONCLUSIONS: HER2/ECD is a prognosis factor that is valuable in evaluating the neoadjuvant treatment efficiency. HER2/ECD also appears to be a helpful surveillance biomarker for the early diagnosis of relapses and to predict the fate of metastases. This study brings evidences to support the use of HER2/ECD in the management of HER2-positive breast cancer.

[Technical recommendations and best practice guidelines for May-Grünwald-Giemsa staining: literature review and insights from the quality assurance]. / Recommandations techniques et règles de bonne pratique pour la coloration de May-Grünwald-Giemsa : revue de la littérature et apport de l'assurance qualité.

May-Grünwald-Giemsa (MGG) stain is a Romanowsky-type, polychromatic stain as those of Giemsa, Leishman and Wright. Apart being the reference method of haematology, it has become a routine stain of diagnostic cytopathology for the study of air-dried preparations (lymph node imprints, centrifuged body fluids and fine needle aspirations). In the context of their actions of promoting the principles of quality assurance in cytopathology, the French Association for Quality Assurance in Anatomic and Cytologic Pathology (AFAQAP) and the French Society of Clinical Cytology (SFCC) conducted a proficiency test on MGG stain in 2013. Results from the test, together with the review of literature data allow pre-analytical and analytical steps of MGG stain to be updated. Recommendations include rapid air-drying of cell preparations/imprints, fixation using either methanol or May-Grünwald alone for 3-10minutes, two-step staining: 50% May-Grünwald in buffer pH 6.8 v/v for 3-5minutes, followed by 10% buffered Giemsa solution for 10-30minutes, and running water for 1-3minutes. Quality evaluation must be performed on red blood cells (RBCs) and leukocytes, not on tumour cells. Under correct pH conditions, RBCs must appear pink-orange (acidophilic) or buff-coloured, neither green nor blue. Leukocyte cytoplasm must be almost transparent, with clearly delineated granules. However, staining may vary somewhat and testing is recommended for automated methods (slide stainers) which remain the standard for reproducibility. Though MGG stain remains the reference stain, Diff-Quik(®) stain can be used for the rapid evaluation of cell samples.

[uPA/PAI-1, Oncotype DX™, MammaPrint(®). Prognosis and predictive values for clinical utility in breast cancer management]. / uPA/PAI-1, Oncotype DX™, Mammaprint(®). Valeurs pronostique et prédictive pour une utilité clinique dans la prise en charge du cancer du sein.

CONTEXT AND AIMS: Breast cancer prognosis and predictive biomarkers development would allow sparing some patients from chemotherapy or identifying patients for whom chemotherapy would be indicated. In this context, in 2009, the French National Cancer Institute, a National Health and Science Agency dedicated to cancer, in collaboration with the French society of senology and breast pathology (SFSPM) published a report on the assessment of the prognostic and the predictive clinical validity of tissular biomarkers, uPA/PAI-1, Oncotype DX™ and MammaPrint(®), in breast cancer management. They concluded that only the uPA/PAI-1 prognosis value reached the highest level of evidence (LOE I according to Hayes 1998 classification). In 2012, it was decided to update this report since new data have emerged and because information disparities among clinicians have been identified. This article aims to present the main conclusions together with the levels of evidence associated with those conclusions. METHODS: The updating process was based on literature published since 2009 appraisal and on multidisciplinary and independent experts' opinion. The levels of evidence (LOE) used are those of the classification defined by Simon in 2009 (updated Hayes 1998 classification): LOE IA and LOE IB: high level of evidence; LOE IIB and LOE IIC: intermediate level of evidence; LOE IIIC and LOE IV-VD: low level of evidence. CONCLUSIONS: Among patients without lymph-node involvement, uPA/PAI-1, invasion process biomarkers, reach the highest level of evidence for 10 years recurrence free survival prognosis (LOE IA according to Simon). The predictive value to anthracyclins chemotherapy remains to be confirmed. Oncotype DX™ and MammaPrint(®) prognosis and predictive value do not reach the LOE I level. This updating' process confirms the 2009 levels of evidence for all the three biomarkers prognosis value. Besides, concerning Oncotype DX™ and MammaPrint(®), new data do not allow to conclude neither to their complementary clinical information to other clinicopathological existing biomarkers nor to a favorable cost-efficiency ratio in therapeutic decision making and this because of the methodological weakness and uncertainty that are identified in the selected studies. Practically, beyond the prognosis and predictive biomarkers validity, the clinical utility of a new biomarker for chemotherapy indication depends on its clinical added information with regard to validated biomarkers (HR, HER2 and Ki67) and to clinicopathological parameters. Since they are the sole validated biomarkers of the invasion process, uPA/PAI-1 could complete clinical information of other clinicopathological factors and consequently could confer an added clinical value. However, data concerning the impact of this information on chemotherapy clinical indication are lacking.

First report of granulomatous mastitis associated with Sjögren's syndrome.

Granulomatous mastitis is a rare and often considered as idiopathic disease. However, clinical examination and thorough diagnostic investigations have to be carried out in order to identify cases that are secondary to infections or systemic diseases since these forms may be cured with appropriate etiologic treatment. To the best of our knowledge, this report is the first to describe the association of granulomatous mastitis with Sjögren's syndrome. We discuss the clinical, pathological and therapeutic implications of this association.

Ki-67: level of evidence and methodological considerations for its role in the clinical management of breast cancer: analytical and critical review.

Clinicians can use biomarkers to guide therapeutic decisions in estrogen receptor positive (ER+) breast cancer. One such biomarker is cellular proliferation as evaluated by Ki-67. This biomarker has been extensively studied and is easily assayed by histopathologists but it is not currently accepted as a standard. This review focuses on its prognostic and predictive value, and on methodological considerations for its measurement and the cut-points used for treatment decision. Data describing study design, patients' characteristics, methods used and results were extracted from papers published between January 1990 and July 2010. In addition, the studies were assessed using the REMARK tool. Ki-67 is an independent prognostic factor for disease-free survival (HR 1.05-1.72) in multivariate analyses studies using samples from randomized clinical trials with secondary central analysis of the biomarker. The level of evidence (LOE) was judged to be I-B with the recently revised definition of Simon. However, standardization of the techniques and scoring methods are needed for the integration of this biomarker in everyday practice. Ki-67 was not found to be predictive for long-term follow-up after chemotherapy. Nevertheless, high KI-67 was found to be associated with immediate pathological complete response in the neoadjuvant setting, with an LOE of II-B. The REMARK score improved over time (with a range of 6-13/20 vs. 10-18/20, before and after 2005, respectively). KI-67 could be considered as a prognostic biomarker for therapeutic decision. It is assessed with a simple assay that could be standardized. However, international guidelines are needed for routine clinical use.

Metabolomic pattern of childhood neuroblastoma obtained by ¹H-high-resolution magic angle spinning (HRMAS) NMR spectroscopy.

BACKGROUND: The aim of this preliminary study is to characterize by ¹H high-resolution magic angle spinning NMR spectroscopy (HRMAS) the metabolic content of intact biopsy samples obtained from 12 patients suffering from neuroblastoma (NB). PROCEDURE: The biochemical NB profile was first compared to normal adrenal medulla. In a second step, the relationship between the tumor metabolic profile and the patients' clinical data was investigated. RESULTS: A higher level of creatine, glutamine/glutamate, acetate and glycine characterized NB biopsies while healthy adrenal medulla tissue contained adrenaline and a larger amount of ascorbic acid. Adrenaline, which was undetectable in NB spectra, represented the metabolic signature of normal adrenal medulla. NB from patients younger than 12 months contained a higher level of acetate and lysine. Conversely, higher amounts of glutathione, glutamate, myo-inositol, glycine, serine and ascorbic acid were detected in NB samples belonging to younger children. Glutamine/glutamate, aspartate, creatine, glycine were characteristic of stage I-II NB. Acetate and creatine were characteristic of stage IV NB. Finally, a relatively higher amount of aspartate, succinate, and glutathione was detected in patients alive without active disease after a mean follow-up of 7 years whereas a higher concentration of acetate and taurine was characteristic of patients with worse prognosis. CONCLUSIONS: Our preliminary results suggest the existence of a complex metabolic reality in NB, probably representative of tumor behavior. However, the real impact of these promising results should be assessed by long-term prospective studies on a larger cohort of patients.

[Update of the GEFPICS' recommendations for HER2 status determination in breast cancers in France]. / Mise à jour des recommandations du GEFPICS pour l'évaluation du statut HER2 dans les cancers du sein en France.

In Europe, patients who may benefit from an HER2 targeted drug are currently selected by immunohistochemistry (IHC). In situ hybridization (ISH) techniques should be used for complementary assessment of ambiguous 2+ IHC cases and for the calibration of the IHC technique. Eligibility to an HER2 target treatment is defined by an HER2 positive status being IHC test 3+ or 2+ amplified. Reliable detection of HER2 status is essential to the appropriate usage of HER2 targeted drugs because its specificity is limited to tumors overexpressing HER2. It is essential that the IHC evaluation of the HER2 status of a mammary carcinoma is optimized and reliable. This GEFPICS' guidelines look over the different steps of the IHC technique, the controls and, the rules for interpretation. Once acquired, this knowledge must be perpetuated by the observation of rules of good technical practice (internal and external controls, quality assurance programs).

Pregnancy and post-partum breast cancer: a prospective study.

BACKGROUND: The concomitant occurrence of breast cancer and pregnancy remains a challenging clinical situation combining ethical and medical problems. There are few prospective data on pregnancy-associated breast cancer (PABC) whose incidence continuously increases. PATIENTS AND METHODS: Forty patients with PABC were compared with 61 non-pregnant, age-matched patients with infiltrative breast carcinomas (BC) diagnosed and followed since 1982. RESULTS: Although PABC and BC tumor size, grade and type, and lymphovascular and lymphnode invasion were similar, the BC cases showed better overall--(p=0.0001) and disease-free (p-0.015) survival. Moreover, the outcome of pregnant patients was worse than post-partum patients (p=0.017). Importantly, the number of PABC patients receiving hormonotherapy was lower than the BC patients (p<0.0004), due to lower estrogen receptor (ER) (p=0.038) and progesterone receptor (PR) (p=0.008) immunohisto-chemical (IH) levels. Retrospective estrogen-regulated pS2/trefoil factor 1 (pS2/TFF1) immunohistochemitry showed no difference between PABC and BC. All the children delivered were healthy. CONCLUSION: Pregnancy and the post-partum period increase breast cancer aggressiveness, pregnancy being the most detrimental. PABC hormone-dependence is under-estimated using ER and PR, and pS2/TFF1 might help in its determination. Appropriate treatment does not impair child outcome.

[Late-onset vaccination-induced subcutaneous pseudolymphoma]. / Pseudolymphome sous-cutané postvaccinal de survenue tardive.

Persistent subcutaneous nodules arise on rare occasions at sites of injection of aluminium hydroxide-adsorbed vaccine. We report a case following a diphtheria, tetanus and pertussis vaccination. The late onset of the lesion, four years after the injection, led to an uncertain preoperative diagnosis. Histopathologic examination showed features of a subcutaneous pseudolymphoma. The demonstration of aluminium by Morin staining and atomic absorption spectrometry on a paraffin-embedded tissue probe supported the diagnosis of a vaccination-induced pseudolymphoma.

[Internal quality control of histological diagnosis in pathology. A nine-year experience]. / Le contrôle qualité interne du diagnostic histologique en anatomie et cytologie pathologiques. A propos d'un vécu de neuf années.

Internal quality control (IQC) is a necessary component of total quality management. We report our experience with an internal audit scheme focusing on the histological diagnosis. We outline other strategies of IQC and analyze the causes of errors and ways to prevent them. Some practical guidelines to initiate this type of procedure are presented. Our audit was designed to check the accuracy of diagnosis, the clarity and completeness of the report, the quality of the documents leading to the diagnosis, and the turn-around time. It consisted of a retrospective analysis of 4185 randomly selected cases (representing 2% of all cases), over nine years. The control took place once a week and was done by two pathologists working as a team. The mean time spent by each pathologist was 45 minutes per week. Errors were scored using a 3-level grading scheme depending on their potential harm or impact on patient care. The overall rate of errors was 1.1%, and 0.1% of errors were potentially harmful to the patients. A single case (0.02%), in which a cancer was missed, had a real impact on patient care. Retrospective analysis of randomly selected cases mirrors the overall activity of a surgical pathology department. Nevertheless, each lab has to develop its own strategy of IQC, based on its size, its functioning, and its objectives. Although it may be difficult to initiate quality assurance when medical time is already limited, it is a helpful procedure in a more and more demanding medical and societal context and a pragmatic step towards "culture of quality".

Sentinel lymph node biopsy and non-sentinel node involvement in special type breast carcinomas with a good prognosis.

This study aimed at identifying factors related to sentinel lymph node (SLN) involvement in patients with tubular, cribriform, mucinous or papillary breast carcinoma and those related to non-SLN metastases if an SLN was positive. Multivariate analyses involved logistic and stepwise regressions. The SLNs harboured metastases in 85 of 572 cases, 78 of whom underwent axillary dissection; 19 presented non-SLN positive disease. Lack of lymphovascular invasion, a tumour size < or = 10 mm and a single SLN removed were the factors predicting an SLN metastasis rate <10%, and patients with these features could be candidates for no surgical axillary staging. A positive SLN proportion of < or = 50% and no lymphovascular invasion were associated with a <10% rate of non-SLN invasion; patients with a positive SLN and these features could be candidates for the omission of completion axillary dissection. The opposite presentation of these factors would mandate SLN biopsy and axillary dissection, respectively.

Sentinel lymph node biopsy in staging small (up to 15 mm) breast carcinomas. Results from a European multi-institutional study.

Sentinel lymph node (SLN) biopsy has become the preferred method for the nodal staging of early breast cancer, but controversy exists regarding its universal use and consequences in small tumors. 2929 cases of breast carcinomas not larger than 15 mm and staged with SLN biopsy with or without axillary dissection were collected from the authors' institutions. The pathology of the SLNs included multilevel hematoxylin and eosin (HE) staining. Cytokeratin immunohistochemistry (IHC) was commonly used for cases negative with HE staining. Variables influencing SLN involvement and non-SLN involvement were studied with logistic regression. Factors that influenced SLN involvement included tumor size, multifocality, grade and age. Small tumors up to 4 mm (including in situ and microinvasive carcinomas) seem to have SLN involvement in less than 10%. Non-SLN metastases were associated with tumor grade, the ratio of involved SLNs and SLN involvement type. Isolated tumor cells were not likely to be associated with further nodal load, whereas micrometastases had some subsets with low risk of non-SLN involvement and subsets with higher proportion of further nodal spread. In situ and microinvasive carcinomas have a very low risk of SLN involvement, therefore, these tumors might not need SLN biopsy for staging, and this may be the approach used for very small invasive carcinomas. If an SLN is involved, isolated tumor cells are rarely if ever associated with non-SLN metastases, and subsets of micrometastatic SLN involvement may be approached similarly. With macrometastases the risk of non-SLN involvement increases, and further axillary treatment should be generally indicated.

Endometrial brush cytology in the surveillance of post-menopausal patients under tamoxifen: a prospective longitudinal study.

OBJECTIVE: To evaluate cytological sampling of endometrium using Endobrush (Lab CCD, Paris, France) in the surveillance of tamoxifen-treated patients. STUDY DESIGN: Between February 1995 and October 1997, 687 tamoxifen-treated patients had serial ultrasound screening for endometrial pathology. In case of endometrial double layer thickness of more than 8mm, a cytological examination of endometrium was attempted followed by hysteroscopy and curettage. RESULTS: One hundred and eighty-nine patients had abnormal endometrial ultrasound findings. Cytological smear was not obtained in 39 patients because of cervical stenosis or pain in 33 and 6 cases, respectively. One hundred and fifty patients had cytological endometrial sampling followed by hysteroscopy and curettage. Cytological and histological findings correlated well in 145 cases (141 benign lesions and 4 endometrial cancers). There were five false positive (four atypia and one cancer). All patients remained free of endometrial cancer at 5 years follow-up. CONCLUSION: In tamoxifen-treated patients, endometrial cytology was reliable for detection of endometrial pathology, and was well accepted by the patients.

Structural abnormalities in islets from very young children with cystic fibrosis may contribute to cystic fibrosis-related diabetes.

Cystic fibrosis (CF)-related diabetes (CFRD) is thought to result from beta-cell injury due in part to pancreas exocrine damage and lipofibrosis. CFRD pancreata exhibit reduced islet density and altered cellular composition. To investigate a possible etiology, we tested the hypothesis that such changes are present in CF pancreata before the development of lipofibrosis. We evaluated pancreas and islet morphology in tissues from very young CF children (<4 years of age), and adult patients with CF and CFRD. The relative number of beta-cells in young CF tissues was reduced by 50% or more when compared to age-matched controls. Furthermore, young CF tissues displayed significantly smaller insulin-positive areas, lower proportion of beta-cells positive for the proliferation marker Ki67 or the ductal marker CK19 vs. control subjects, and islet inflammatory cell infiltrates, independently of the severity of the exocrine lesion and in the absence of amyloid deposits. CFRD pancreata exhibited greater islet injury with further reduction in islet density, decreased relative beta-cell number, and presence of amyloid deposits. Together, these results strongly suggest that an early deficiency in beta-cell number in infants with CF may contribute to the development of glucose intolerance in the CF pediatric population, and to CFRD, later in life.

Adequately defining tumor cell proportion in tissue samples for molecular testing improves interobserver reproducibility of its assessment.

Gene mutation status assessment of tumors has become standard practice in diagnostic pathology. This is done using samples comprising tumor cells but also non-tumor cells, which may dramatically dilute the proportion of tumor DNA and induce false negative results. Increasing sensitivity of molecular tests presently allows detection of a targeted mutation in a sample with a small percentage of tumor cells, but assessment of tumor cellularity remains essential to adequately interpret the results of molecular tests. Comprehensive tumor cell counting would provide the most reliable approach but is time consuming, and therefore rough global estimations are used, the reliability of which has been questioned in view of their potential clinical impact. The French association for quality assurance in pathology (AFAQAP) conducted two external quality assurance schemes, partly in partnership with the French group of oncology cytogenomics (GFCO). The purpose of the schemes was to (1) evaluate how tumor cellularity is assessed on tissue samples, (2) identify reasons for discrepancies, and (3) provide recommendations for standardization and improvement. Tumor cell percentages in tissue samples of lung and colon cancer were estimated by 40-50 participants, on 10 H&E virtual slides and 20 H&E conventional slides. The average difference between lowest and highest estimated percentage was 66. This was largely due to inadequate definition of cellularity, reflecting confusion between the percentage of tumor cells and the percentage of the area occupied by tumor in the assessed region. The widest range of interobserver variation was observed for samples with dense or scattered lymphocytic infiltrates or with mucinous stroma. Estimations were more accurate in cases with a low percentage of tumor cells. Macrodissection of the most homogeneous area in the tissue reduced inter-laboratory variation. We developed a rating system indicating potential clinical impact of a discrepancy. Fewer discrepancies were clinically relevant since the study was conducted. Although semi-quantitative estimations remain somewhat subjective, their reliability improves when tumor cellularity is adequately defined and heterogeneous tissue samples are macrodissected for molecular analysis.