Joint Region and Nucleus Segmentation for Characterization of Tumor Infiltrating Lymphocytes in Breast Cancer.

Histologic assessment of stromal tumor infiltrating lymphocytes (sTIL) as a surrogate of the host immune response has been shown to be prognostic and potentially chemo-predictive in triple-negative and HER2-positive breast cancers. The current practice of manual assessment is prone to intra- and inter-observer variability. Furthermore, the inter-play of sTILs, tumor cells, other microenvironment mediators, their spatial relationships, quantity, and other image-based features have yet to be determined exhaustively and systemically. Towards analysis of these aspects, we developed a deep learning based method for joint region-level and nucleus-level segmentation and classification of breast cancer H&E tissue whole slide images. Our proposed method simultaneously identifies tumor, fibroblast, and lymphocyte nuclei, along with key histologic region compartments including tumor and stroma. We also show how the resultant segmentation masks can be combined with seeding approaches to yield accurate nucleus classifications. Furthermore, we outline a simple workflow for calibrating computational scores to human scores for consistency. The pipeline identifies key compartments with high accuracy (Dice= overall: 0.78, tumor: 0.83, and fibroblasts: 0.77). ROC AUC for nucleus classification is high at 0.89 (micro-average), 0.89 (lymphocytes), 0.90 (tumor), and 0.78 (fibroblasts). Spearman correlation between computational sTIL and pathologist consensus is high (R=0.73, p<0.001) and is higher than inter-pathologist correlation (R=0.66, p<0.001). Both manual and computational sTIL scores successfully stratify patients by clinical progression outcomes.

Sebastian syndrome with abnormal platelet response to ristocetin.

BACKGROUND: Sebastian syndrome is characterized by enlarged platelets and Döhle-like body leukocyte inclusions. This syndrome is an MYH-9-related disease, a group that also includes May-Hegglin anomaly and Fechtner syndrome. The differential diagnosis of the MYH-9 diseases requires ultrastructural studies. Certain in vitro aggregation responses may be abnormal in these conditions. OBSERVATIONS: A 6-month-old boy presented with macrothrombocytopenia but no overt bleeding tendency. Giant platelets and Döhle-like body leukocyte inclusions were present in blood smears from both the patient and his mother. Electron microscopy confirmed ultrastructural features consistent with Sebastian syndrome. Platelet aggregation studies were normal except for an impaired response to the agonist ristocetin. CONCLUSIONS: In this patient peripheral blood analyses and platelet aggregation studies revealed disease features shared with the Bernard-Soulier syndrome, but this syndrome was excluded by cellsurface glycoprotein analysis.

The impact of assessing specimen adequacy and number of needle passes for fine-needle aspiration biopsy of thyroid nodules.

Fine-needle aspiration biopsy (FNAB) of thyroid nodules is a safe, cost-effective procedure but the rates of inadequate cytology specimens range from approximately 1% to 15%. This study tests the hypothesis that ultrasonographically (US) guided FNAB and onsite assessment of cytology improves the adequacy rate of FNAB. A retrospective analysis was performed on 693 thyroid FNAB specimens obtained with and without ultrasound guidance and with or without onsite cytology assessment. Overall, 29 specimens (4%) were inadequate for diagnosis. Among 163 cystic nodules and 530 solid nodules, inadequacy rates were 15% (n = 24) and 1% (n = 5) respectively (p = 0.0001). An onsite assessment of cytology for adequacy was done in 550 cases (83%), which was more accurately performed by a cytopathologist (97%) than a cytotechnologist (93%, p = 0.015). With US-guided FNAB, 3% of the cytology specimens were inadequate, compared to a 7% rate when US was not done (p = 0.003). The mean number of needle punctures necessary for an adequate specimen was 3.8 +/- 0.06 (median, 3.0; range, 1-11), which was different among various types of doctors, ranging from 3.2 +/- 0.07 to 5.4 +/- 0.12 (p = 0.001 analysis of variance [ANOVA]). The fewest number of needle passes to achieve an adequate specimen were required by university endocrinologists and pathologists working together (average, 3.2 +/- 0.07; median, 3.0; range, 1-11). Sample inadequacy rate varied significantly among physician groups, ranging from 3% to 18% (p = 0.0001 ANOVA). Stepwise regression analysis showed that onsite assessment of cytology, US-guided FNAB (p = 0.16), and cystic nature of the nodule (p < 0.0001 for all) correlated with adequacy of the specimen. We conclude that US-guided FNAB with onsite evaluation of cytology specimens substantially increases the adequacy of cytology specimens and decreases the number of required needle passes, which ultimately reduces patient discomfort and diagnostic errors, thus raising the question as to whether this should eventually become the standard of care. We believe this is a goal that training programs should strive to achieve.

The utility of p16(Ink4a) in discriminating between cervical intraepithelial neoplasia 1 and nonneoplastic equivocal lesions of the cervix.

CONTEXT: The protein p16(Ink4a) is overexpressed in cervical lesions associated with high-risk human papillomavirus (HPV) subtypes 16 and 18, but not in low-risk HPV subtypes 6 and 11 or non-HPV-associated cervical lesions. OBJECTIVE: To determine whether p16(Ink4a) expression in equivocal cervical lesions helps distinguish atypical non-HPV changes from HPV-related changes. DESIGN: One hundred ninety-one cervical lesions, including 81 cervical intraepithelial neoplasia 1, 52 squamous metaplasia, 33 cellular features suggestive of HPV-related change, 9 reserve cell hyperplasia, 4 microglandular hyperplasia, and 12 inflammatory cervicitis, were randomly selected from archival cervical biopsy specimens. All 191 samples were studied with p16(Ink4a) (JC8 monoclonal antibody). Reactivity for p16(Ink4a) was scored on a 3-tier system as follows: negative, 0% to 5% cells reactive; focal/scattered positive, greater than 5% and less than or equal to 80% cells reactive; diffuse positive, greater than 80% cells reactive. Reactivity was based on normal/reactive cervical specimens where anti-p16 antibody was negative/weakly expressed in non-cervical epithelial cells. Cervical intraepithelial neoplasia 1 lesions not reactive for p16(Ink4a) were investigated for the presence of high-risk HPV by real-time polymerase chain reaction. RESULTS: No p16(Ink4a) reactivity was detected in the cervical lesions associated with atypical non-HPV change. Eleven of the cervical intraepithelial neoplasia 1 lesions showed focal/scattered reactivity expression for p16(Ink4a), and 19 of the CIN 1 lesions had diffuse reactivity. Fifty of 51 of the CIN 1 lesions negative for p16(Ink4a) were real-time polymerase chain reaction negative for the presence of high-risk HPV; 1 was real-time polymerase chain reaction positive for high-risk HPV. CONCLUSIONS: The data support the routine use of p16(Ink4a) immunohistochemical evaluation of cervical biopsy specimens for better discrimination of non-HPV-associated lesions from HPV-related lesions.

Validation of a five-tier cytodiagnostic system for thyroid fine needle aspiration biopsies using cytohistologic correlation.

OBJECTIVE: Approximately 70-80% of thyroid fine needle aspiration biopsies (FNAB) can distinguish benign from malignant thyroid nodules. However, much interpretive diagnostic difficulty arises with the remaining 20-30% of cases. These problematic thyroid aspirations have been placed in various diagnostic categories, which collectively have led to confusion and a negative impact on the clinical management of patients with thyroid nodules. We present our experience using a five-tier system, including the diagnostic terminology: benign, indeterminate, suspicious, malignant, and unsatisfactory. DESIGN: Thyroid FNABs were diagnosed using a five-tier system from 200 consecutive patients and the subsequent surgical excisions were correlated. MAIN OUTCOME: Overall, there was an excellent association between the five diagnostic categories and predicting benign versus neoplastic thyroid nodules (LR = 96.06, X(2) = 76.49, and phi = 0.618, df = 4, p < 0.0001). A negative cytologic diagnosis carries a negative predictive value of 92%, while an indeterminate, suspicious, and malignant cytologic diagnosis carries a positive predictive value of 50, 71, and 100% respectively. The estimated sensitivity for an indeterminate, suspicious, or malignant cytologic diagnosis varied from 70 to 89%, while the specificity increased from 57 to 92 to 100%, respectively. CONCLUSIONS: The data presented shows that the five diagnostic categories of thyroid FNAB are excellent at distinguishing benign from neoplastic thyroid nodules. Both the indeterminate and suspicious categories, while not statistically different from each other in predicting benign from neoplasia, are statistically different from obviously benign and obviously malignant categories. These results support the need for an indeterminate and/or suspicious category.

Perceptions of diagnostic terminology and cytopathologic reporting of fine-needle aspiration biopsies of thyroid nodules: a survey of clinicians and pathologists.

INTRODUCTION: Guidelines on thyroid fine-needle aspiration biopsy (FNABs) reporting calls for unambiguous diagnostic terminology in order to maximize treatment. This study evaluates how pathologists follow the guidelines and clinicians understand the diagnostic categories in terms of patient care. DESIGN: Survey 1 asked pathologists who perform/interpret FNABs which of "atypical," "indeterminate," "suspicious," and "nondiagnostic" they routinely use. Survey 2 asked clinicians who treat thyroid nodules to correlate these categories to the options of "negative FNAB/follow-up," "repeat FNAB," and "proceed to surgery." The anonymous, voluntary results were entered into a database and analyzed. MAIN OUTCOME: Pathologists' and clinicians' response rates were 70% and 35%, respectively. Survey 1: 27% of pathologists used three, 27% used one, and 44% used two categories. Survey 2: 98% clinicians would repeat the FNAB with a "nondiagnostic" and 96% opted for surgery with a "suspicious" diagnosis. "Indeterminate" prompted 58% to repeat the FNAB and 32% to send the patient to surgery. "Atypical" would lead 37% to repeat the FNAB and 52% to send the patient to surgery. CONCLUSIONS: Pathologists actively use the terminology "suspicious," "indeterminate," or "atypical," which cause confusion in some clinicians. These results support the need for a more standardized terminology for FNAB reporting and education of the clinicians on that terminology.

Uterine-like mass with features of an extrauterine adenomyoma presenting 22 years after total abdominal hysterectomy-bilateral salpingo-oophorectomy: a case report and review of the literature.

Adenomyoma is a benign tumor composed of smooth muscle and benign endometrium. These tumors typically originate within the uterus. An extrauterine adenomyoma is a rare entity. We report a uterine-like mass consistent with an extrauterine adenomyoma presenting 22 years following a total abdominal hysterectomy and bilateral salpingo-oophorectomy. The mass was pear-shaped with uterine-type smooth muscle and a cavity lined by functional endometrial glands and stroma. To our knowledge, only 4 other cases of an extrauterine uterine-like mass are reported in the literature. Three involved the ovary, while one was located adjacent to the broad ligament with normal pelvic organs. Although none of these other uterus-like masses were described as adenomyomas with uterine-like features, the histologic findings are strikingly similar. An understanding of the müllerian system suggests that either an embryologic malformation or a differential multipotentiality existing in the subcoelomic tissues in response to hormonal stimulation results in a supernumerary müllerian structure like a uterus, as observed in this case. The presence of endometrial glands and stroma in the mass confirms that the tissues in this mass are hormonally responsive. It is most likely that this uterine-like mass arose from the tissues of the secondary müllerian system in response to estrogenic stimulation.

Association between invasive squamous cell carcinoma of the vulva and ABO blood group.

OBJECTIVE: To determine whether the distribution of ABO blood group among women with invasive squamous cell carcinoma (SCC) of the vulva is different from that among a population of women treated for nonneoplastic gynecologic disease. METHODS: A retrospective analysis of pathology reports and blood bank records from January 1996 through September 2003 was performed. The distribution of ABO blood group for 33 women diagnosed with invasive SCC of the vulva was determined. ABO blood group was also recorded for 100 female patients (controls) who underwent a gynecologic procedure for a nonneoplastic process during the period January 2003 through November 2003. The blood group phenotype distribution for the study groups and the controls was compared by an incidence ratio. RESULTS: Statistical analysis gave an incidence ratio for blood group types A and O of 1.55 (p < .20) when the patients with invasive SCC of the vulva were compared with the controls. The p value was not significant. Similarly, the incidence ratio for blood group types B and O equaled 0.386 (p = 1). Again the p value was not significant. In fact, none of the incidence ratios calculated were statistically significant. Although not statistically significant, the incidence ratio for the invasive vulvar SCC group was >1. This may indicate that there is a trend for women with invasive SCC of the vulva to have blood group type A. CONCLUSIONS: Results of this study do not suggest an association between blood group type A, or any other blood group, and vulvar SCC. This study was similar in patient size to another study (33 patients with vulvar SCC vs. 39 patients with vulvar SCC, respectively). The disparity of the results in determining the significance of blood group type A as an associated factor for vulvar SCC in this and the other study may be due to the limited size of the study populations. Additional investigation is needed to further evaluate this issue.