Automatic cellularity assessment from post-treated breast surgical specimens.

Neoadjuvant treatment (NAT) of breast cancer (BCa) is an option for patients with the locally advanced disease. It has been compared with standard adjuvant therapy with the aim of improving prognosis and surgical outcome. Moreover, the response of the tumor to the therapy provides useful information for patient management. The pathological examination of the tissue sections after surgery is the gold-standard to estimate the residual tumor and the assessment of cellularity is an important component of tumor burden assessment. In the current clinical practice, tumor cellularity is manually estimated by pathologists on hematoxylin and eosin (H&E) stained slides, the quality, and reliability of which might be impaired by inter-observer variability which potentially affects prognostic power assessment in NAT trials. This procedure is also qualitative and time-consuming. In this paper, we describe a method of automatically assessing cellularity. A pipeline to automatically segment nuclei figures and estimate residual cancer cellularity from within patches and whole slide images (WSIs) of BCa was developed. We have compared the performance of our proposed pipeline in estimating residual cancer cellularity with that of two expert pathologists. We found an intra-class agreement coefficient (ICC) of 0.89 (95% CI of [0.70, 0.95]) between pathologists, 0.74 (95% CI of [0.70, 0.77]) between pathologist #1 and proposed method, and 0.75 (95% CI of [0.71, 0.79]) between pathologist #2 and proposed method. We have also successfully applied our proposed technique on a WSI to locate areas with high concentration of residual cancer. The main advantage of our approach is that it is fully automatic and can be used to find areas with high cellularity in WSIs. This provides a first step in developing an automatic technique for post-NAT tumor response assessment from pathology slides. © 2017 International Society for Advancement of Cytometry.

Cutaneous melanoma in 1-year-old child: an insight on infantile melanoma.

In the past, malignant melanoma (MM) is a diagnosis of unheard in children, but nowadays MM is a very rare malignancy in children. Its diagnosis requires careful interpretation of the pathological diagnostic criteria with clinical correlation of the findings. In this study, the authors are presenting a pigmented lesion in a 12-month-old girl, which was present since her birth with increase in size and shape. The authors discussed the difficulty that confronted them in making a diagnosis of MM and the differential diagnosis.

Automated and Manual Quantification of Tumour Cellularity in Digital Slides for Tumour Burden Assessment.

The residual cancer burden index is an important quantitative measure used for assessing treatment response following neoadjuvant therapy for breast cancer. It has shown to be predictive of overall survival and is composed of two key metrics: qualitative assessment of lymph nodes and the percentage of invasive or in situ tumour cellularity (TC) in the tumour bed (TB). Currently, TC is assessed through eye-balling of routine histopathology slides estimating the proportion of tumour cells within the TB. With the advances in production of digitized slides and increasing availability of slide scanners in pathology laboratories, there is potential to measure TC using automated algorithms with greater precision and accuracy. We describe two methods for automated TC scoring: 1) a traditional approach to image analysis development whereby we mimic the pathologists' workflow, and 2) a recent development in artificial intelligence in which features are learned automatically in deep neural networks using image data alone. We show strong agreements between automated and manual analysis of digital slides. Agreements between our trained deep neural networks and experts in this study (0.82) approach the inter-rater agreements between pathologists (0.89). We also reveal properties that are captured when we apply deep neural network to whole slide images, and discuss the potential of using such visualisations to improve upon TC assessment in the future.

A Cluster-then-label Semi-supervised Learning Approach for Pathology Image Classification.

Completely labeled pathology datasets are often challenging and time-consuming to obtain. Semi-supervised learning (SSL) methods are able to learn from fewer labeled data points with the help of a large number of unlabeled data points. In this paper, we investigated the possibility of using clustering analysis to identify the underlying structure of the data space for SSL. A cluster-then-label method was proposed to identify high-density regions in the data space which were then used to help a supervised SVM in finding the decision boundary. We have compared our method with other supervised and semi-supervised state-of-the-art techniques using two different classification tasks applied to breast pathology datasets. We found that compared with other state-of-the-art supervised and semi-supervised methods, our SSL method is able to improve classification performance when a limited number of labeled data instances are made available. We also showed that it is important to examine the underlying distribution of the data space before applying SSL techniques to ensure semi-supervised learning assumptions are not violated by the data.

Assessment of the Intraoperative Consultation Service Rendered by General Pathologists in a Scenario Where a Well-Defined Decision Algorithm Is Followed.

OBJECTIVES: Intraoperative consultation (IOC) remains an area of general practice even within subspecialized pathology departments. This study assesses the IOCs rendered in a general pathology setting where surgeons integrate these results in a well-defined algorithm, developed with the input of specialized pathologists. METHODS: The surgical decisions to perform lymphadenectomy in patients with endometrial adenocarcinoma operated on at our institution between January 2003 and June 2015 as a result of the IOC assessment of tumor size, histologic grade, and depth of invasion in the hysterectomy specimen were analyzed. RESULTS: Frozen section (FS) was examined in 801 cases. In comparison to permanent section analysis, FS International Federation of Gynecology and Obstetrics (FIGO) grade had an overall accuracy of 0.95 (95% confidence interval [CI], 0.93-0.98). The FS depth of invasion had an overall accuracy of 0.92 (95% CI, 0.89-0.94). FIGO grade was not documented in 47.8%, the depth of myometrial invasion in 45.2%, and tumor size in 41.8% of the pathology reports. CONCLUSIONS: The high omission rate of the needed parameters by the general pathologists would question their overall understanding of the paradigm shift intended by this algorithm. Possible explanations of this phenomenon and potential solutions are discussed.

Comprehensive survey of HRAS, KRAS, and NRAS mutations in proliferative thyroid lesions from an ethnically diverse population.

BACKGROUND: The distribution and kind of rat sarcoma viral oncogenes homolog (RAS) mutations, as well as their clinical impact on different types of thyroid lesions, vary widely among the different populations studied. We performed a comprehensive mutational survey in the highly related RAS genes HRAS, KRAS, and NRAS in a case series of proliferative thyroid lesions with known BRAF mutational status, originating from an ethnically diverse group. MATERIALS AND METHODS: Mutational hotspot regions encompassing codons 12, 13, and 61 of the RAS genes were directly sequenced in 381 cases of thyroid lesions. In addition, the putative NRAS hotspot region encompassing codon 97 was sequenced in 36 thyroid lesions. The case series included lesions of Hashimoto's thyroiditis (HT), nodular goiters, hyperplastic nodules, follicular adenomas (FAs), Hurthle cell variants of FA, papillary thyroid carcinomas (PTCs), follicular variants of PTC (FVPTCs), microcarcinomas of PTC (micro PTCs; tumor size ≤1 cm), follicular TCs (FTCs), Hurthle cell variants of FTC, and non-well-differentiated TCs (NWDTCs). RESULTS: We identified RAS mutations in 16 out of 57 (28.1%) FAs, 2 out of 8 (25%) NWDTCs, 8 out of 42 (19.0%) FVPTCs, 2 out of 10 (20.0%) FTCs, 1 out of 12 (8.3%) Hurthle cell variants of FA, 3 out of 46 (6.5%) goiters, 1 out of 18 (5.6%) hyperplastic nodules, 3 out of 56 (5.4%) micro PTCs, 2 out of 115 (1.7%) PTCs, 0 out of 7 (0%) Hurthle cell variants of FTC, and 0 out of 10 (0%) HT lesions. NRAS codon 61 mutation was the predominant form, followed by HRAS codon 61 mutation. Only three mutations affected RAS codons 12 and 13, two of which were identified in goiters. No codon 97 mutation was detected in the examined FVPTCs. An as yet undescribed deletion of KRAS codon 59 was identified in one FA. DISCUSSION: RAS mutations in our case series were commonly associated with follicular-patterned thyroid lesions. Our data suggest that FAs with a RAS mutation may constitute precursor lesions for TC with follicular histology. The newly-discovered KRAS codon 59 deletion is one of the first reported codon deletions in a RAS hotspot region.

BRAF mutations in thyroid tumors from an ethnically diverse group.

BACKGROUND: The molecular etiology of thyroid carcinoma (TC) and other thyroid diseases which may present malignant precursor lesions is not fully explored yet. The purpose of this study was to estimate frequency, type and clinicopathological value of BRAF exon 15 mutations in different types of cancerous and non-cancerous thyroid lesions originating in an ethnically diverse population. METHODS: BRAF exon 15 was sequenced in 381 cases of thyroid lesions including Hashimoto´s thyroiditis, nodular goiters, hyperplastic nodules, follicular adenomas (FA), papillary TC (PTC), follicular variant PTC (FVPTC), microcarcinomas of PTC (micro PTC; tumor size ≤ 1 cm), follicular TC (FTC), and non-well differentiated TC (non-WDTC). RESULTS: We identified BRAF mutations in one of 69 FA, 72 of 115 (63%) PTC, seven of 42 (17%) FVPTC, 10 of 56 (18%) micro PTC, one of 17 (6%) FTC, and one of eight (13%) non-WDTC. Most of the cases showed the common V600E mutation. One case each of PTC, FVPTC, and FTC harbored a K601E mutation. A novel BRAF mutation was identified in a FA leading to deletion of threonine at codon 599 (p.T599del). A rare 3-base pair insertion was detected in a stage III PTC resulting in duplication of threonine at codon 599 (p.T599dup). Patients with PTC harboring no BRAF mutation (BRAFwt) were on average younger than those with a BRAF mutation (BRAFmut) in the PTC (36.6 years vs. 43.8 years). Older age (≥ 45 years) in patients with PTC was significantly associated with tumor size ≥ 4 cm (P = 0.018), vessel invasion (P = 0.004), and distant metastasis (P = 0.001). Lymph node (LN) involvement in PTC significantly correlated with tumor size (P = 0.044), and vessel invasion (P = 0.013). Of notice, taken the whole TC group, family history of thyroid disease positively correlated with capsular invasion (P = 0.025). CONCLUSIONS: Older age is manifold associated with unfavorable tumor markers in our series. The K601E identified in a PTC, FVPTC, and FTC seems to be more distributed among different histological types of TC than previously thought. The T599del is a yet undescribed mutation and the rare T599dup has not been reported as a mutation in PTC so far.

Mutational screening of RET, HRAS, KRAS, NRAS, BRAF, AKT1, and CTNNB1 in medullary thyroid carcinoma.

BACKGROUND: Screening medullary thyroid carcinomas (MTCs) for rearranged during transfection (RET) mutations becomes increasingly important for clinical assessment of the disease. The role of mutations in other genes including RAS (i.e. HRAS, KRAS, and NRAS), v-raf murine sarcoma viral oncogene homolog B1 (BRAF), v-akt murine thymoma viral oncogene homolog 1 (AKT1), and CTNNB1 (ß-catenin) is unknown or not fully explored yet for this disease. MATERIALS AND METHODS: Formalin-fixed and paraffin-embedded (FFPE) material was the primary source for screening 13 sporadic and inherited MTCs and matched non-tumor specimens. Multiplex PCR was included in the PCR protocol. Sequence analysis encompassed mutational hotspot regions in RET exons 5, 8, 10, 11, and 13 to 16; HRAS exons 1 and 2; KRAS exons 1 and 2; NRAS exons 1 and 2; BRAF exon 15; AKT1 exon 2, and CTNNB1 exon 3. RESULTS: We identified RET mutations in seven of 13 MTCs: five RET-positive cases revealed a mutation in exon 16 (M918T) and two a mutation in exon 10 (C618S and C620S). In four of the RET-positive cases, the mutation was inherited, out of which three were reportedly associated with a multiple endocrine neoplasia type 2 (MEN2) syndrome, i.e. MEN2A (C618S), MEN2A/familial MTC (FMTC) (C620S), and MEN2B (M918T). These cases reflect the known MEN2 genotype-phenotype correlation. Three of the five stage IVc MTCs were inherited RET-positive cases. Mutational screening in HRAS, KRAS, NRAS, BRAF, AKT1, and CTNNB1 disclosed one sporadic RET-negative MTC (stage III) with mutation in HRAS codon 13 (G13R). CONCLUSION: Our study supports the clinical relevance of screening MTC patients for RET mutations. The role of RAS mutations, in particular HRAS mutations, in sporadic RET-negative MTC has not been fully explored yet. Mutations in BRAF, AKT1, and CTNNB1 are likely not to play a role in MTC.