Opening the Black Box: Spatial Transcriptomics and the Relevance of Artificial Intelligence-Detected Prognostic Regions in High-Grade Serous Carcinoma.

Image-based deep learning models are used to extract new information from standard hematoxylin and eosin pathology slides; however, biological interpretation of the features detected by artificial intelligence (AI) remains a challenge. High-grade serous carcinoma of the ovary (HGSC) is characterized by aggressive behavior and chemotherapy resistance, but also exhibits striking variability in outcome. Our understanding of this disease is limited, partly due to considerable tumor heterogeneity. We previously trained an AI model to identify HGSC tumor regions that are highly associated with outcome status but are indistinguishable by conventional morphologic methods. Here, we applied spatially resolved transcriptomics to further profile the AI-identified tumor regions in 16 patients (8 per outcome group) and identify molecular features related to disease outcome in patients who underwent primary debulking surgery and platinum-based chemotherapy. We examined formalin-fixed paraffin-embedded tissue from (1) regions identified by the AI model as highly associated with short or extended chemotherapy response, and (2) background tumor regions (not identified by the AI model as highly associated with outcome status) from the same tumors. We show that the transcriptomic profiles of AI-identified regions are more distinct than background regions from the same tumors, are superior in predicting outcome, and differ in several pathways including those associated with chemoresistance in HGSC. Further, we find that poor outcome and good outcome regions are enriched by different tumor subpopulations, suggesting distinctive interaction patterns. In summary, our work presents proof of concept that AI-guided spatial transcriptomic analysis improves recognition of biologic features relevant to patient outcomes.

The 1000 Mitoses Project: A Consensus-Based International Collaborative Study on Mitotic Figures Classification.

Introduction. The identification of mitotic figures is essential for the diagnosis, grading, and classification of various different tumors. Despite its importance, there is a paucity of literature reporting the consistency in interpreting mitotic figures among pathologists. This study leverages publicly accessible datasets and social media to recruit an international group of pathologists to score an image database of more than 1000 mitotic figures collectively. Materials and Methods. Pathologists were instructed to randomly select a digital slide from The Cancer Genome Atlas (TCGA) datasets and annotate 10-20 mitotic figures within a 2 mm2 area. The first 1010 submitted mitotic figures were used to create an image dataset, with each figure transformed into an individual tile at 40x magnification. The dataset was redistributed to all pathologists to review and determine whether each tile constituted a mitotic figure. Results. Overall pathologists had a median agreement rate of 80.2% (range 42.0%-95.7%). Individual mitotic figure tiles had a median agreement rate of 87.1% and a fair inter-rater agreement across all tiles (kappa = 0.284). Mitotic figures in prometaphase had lower percentage agreement rates compared to other phases of mitosis. Conclusion. This dataset stands as the largest international consensus study for mitotic figures to date and can be utilized as a training set for future studies. The agreement range reflects a spectrum of criteria that pathologists use to decide what constitutes a mitotic figure, which may have potential implications in tumor diagnostics and clinical management.

Increased MIB-1 expression in salivary gland pleomorphic adenoma that recurs and undergoes malignant transformation.

The objective of this retrospective study was to explore possible changes in histopathological features and expression of cyclin D1 and MIB-1 in salivary gland pleomorphic adenoma (PA) that recur or undergo malignant transformation. Knowledge of these characteristics might help to guide the management of these rare tumors. The histopathology and immunohistochemical staining characteristics of such tumors were analyzed in a cohort of 65 patients constituting three different groups of tumors: PA, recurrent pleomorphic adenoma (RPA) and carcinoma ex PA (CxPA). The RPAs were divided into two subgroups: primary PA that were known to recur later (PA-prim) and recurrent tumors appearing after a primary tumor (PA-rec). RPAs and CxPAs were compared with PAs without recurrence, which served as a control group. In our study, CxPA and PA-rec, but not PA-prim, showed increased MIB-1 expression compared with the control group. Neither cyclin D1 expression nor any histopathological features showed any association in statistical analyses. CxPA showed increased mitotic activity, squamous metaplasia, and nuclear atypia. Tumor multifocality was more frequent in PA-rec and CxPA. The different MIB-1 expression in CxPA and PA-rec in comparison to PA-prim suggests that the changes in expression could develop after the primary tumor.

QuantISH: RNA in situ hybridization image analysis framework for quantifying cell type-specific target RNA expression and variability.

RNA in situ hybridization (RNA-ISH) is a powerful spatial transcriptomics technology to characterize target RNA abundance and localization in individual cells. This allows analysis of tumor heterogeneity and expression localization, which are not readily obtainable through transcriptomic data analysis. RNA-ISH experiments produce large amounts of data and there is a need for automated analysis methods. Here we present QuantISH, a comprehensive open-source RNA-ISH image analysis pipeline that quantifies marker expressions in individual carcinoma, immune, and stromal cells on chromogenic or fluorescent in situ hybridization images. QuantISH is designed to be modular and can be adapted to various image and sample types and staining protocols. We show that in chromogenic RNA in situ hybridization images of high-grade serous carcinoma (HGSC) QuantISH cancer cell classification has high precision, and signal expression quantification is in line with visual assessment. We further demonstrate the power of QuantISH by showing that CCNE1 average expression and DDIT3 expression variability, as captured by the variability factor developed herein, act as candidate biomarkers in HGSC. Altogether, our results demonstrate that QuantISH can quantify RNA expression levels and their variability in carcinoma cells, and thus paves the way to utilize RNA-ISH technology.

Comparative Proteomic Analysis of HPV(+) Oropharyngeal Squamous Cell Carcinoma Recurrence.

Deintensification therapy for human papillomavirus-related oropharyngeal squamous cell carcinoma (HPV(+) OPSCC) is under active investigation. An adaptive treatment approach based on molecular stratification could identify high-risk patients predisposed to recurrence and better select for appropriate treatment regimens. Collectively, 40 HPV(+) OPSCC FFPE samples (20 disease-free, 20 recurrent) were surveyed using mass spectrometry-based proteomic analysis via data-independent acquisition to obtain fold change and false discovery differences. Ten-year overall survival was 100.0 and 27.7% for HPV(+) disease-free and recurrent cohorts, respectively. Of 1414 quantified proteins, 77 demonstrated significant differential expression. Top enriched functional pathways included those involved in programmed cell death (73 proteins, p = 7.43 × 10-30), apoptosis (73 proteins, p = 5.56 × 10-9), ß-catenin independent WNT signaling (47 proteins, p = 1.45 × 10-15), and Rho GTPase signaling (69 proteins, p = 1.09 × 10-5). PFN1 (p = 1.0 × 10-3), RAD23B (p = 2.9 × 10-4), LDHB (p = 1.0 × 10-3), and HINT1 (p = 3.8 × 10-3) pathways were significantly downregulated in the recurrent cohort. On functional validation via immunohistochemistry (IHC) staining, 46.9% (PFN1), 71.9% (RAD23B), 59.4% (LDHB), and 84.4% (HINT1) of cases were corroborated with mass spectrometry findings. Development of a multilateral molecular signature incorporating these targets may characterize high-risk disease, predict treatment response, and augment current management paradigms in head and neck cancer.

Artificial intelligence-based image analysis can predict outcome in high-grade serous carcinoma via histology alone.

High-grade extrauterine serous carcinoma (HGSC) is an aggressive tumor with high rates of recurrence, frequent chemotherapy resistance, and overall 5-year survival of less than 50%. Beyond determining and confirming the diagnosis itself, pathologist review of histologic slides provides no prognostic or predictive information, which is in sharp contrast to almost all other carcinoma types. Deep-learning based image analysis has recently been able to predict outcome and/or identify morphology-based representations of underlying molecular alterations in other tumor types, such as colorectal carcinoma, lung carcinoma, breast carcinoma, and melanoma. Using a carefully stratified HGSC patient cohort consisting of women (n = 30) with similar presentations who experienced very different treatment responses (platinum free intervals of either ≤ 6 months or ≥ 18 months), we used whole slide images (WSI, n = 205) to train a convolutional neural network. The neural network was trained, in three steps, to identify morphologic regions (digital biomarkers) that are highly associating with one or the other treatment response group. We tested the classifier using a separate 22 slide test set, and 18/22 slides were correctly classified. We show that a neural network based approach can discriminate extremes in patient response to primary platinum-based chemotherapy with high sensitivity (73%) and specificity (91%). These proof-of-concept results are novel, because for the first time, prospective prognostic information is identified specifically within HGSC tumor morphology.

Quantitative survival impact of composite treatment delays in head and neck cancer.

BACKGROUND: Multidisciplinary management of head and neck cancer (HNC) must reconcile increasingly sophisticated subspecialty care with timeliness of care. Prior studies examined the individual effects of delays in diagnosis-to-treatment interval, postoperative interval, and radiation interval but did not consider them collectively. The objective of the current study was to investigate the combined impact of these interwoven intervals on patients with HNC. METHODS: Patients with HNC who underwent curative-intent surgery with radiation were identified in the National Cancer Database between 2004 and 2013. Multivariable models were constructed using restricted cubic splines to determine nonlinear relations with overall survival. RESULTS: Overall, 15,064 patients were evaluated. After adjustment for covariates, only prolonged postoperative interval (P < .001) and radiation interval (P < .001) independently predicted for worse outcomes, whereas the association of diagnosis-to-treatment interval with survival disappeared. By using multivariable restricted cubic spline functions, increasing postoperative interval did not affect mortality until 40 days after surgery, and each day of delay beyond this increased the risk of mortality until 70 days after surgery (hazard ratio, 1.14; 95% confidence interval, 1.01-1.28; P = .029). For radiation interval, mortality escalated continuously with each additional day of delay, plateauing at 55 days (hazard ratio, 1.25; 95% confidence interval, 1.11-1.41; P < .001). Delays beyond these change points were not associated with further survival decrements. CONCLUSIONS: Increasing delays in postoperative and radiation intervals are associated independently with an escalating risk of mortality that plateaus beyond certain thresholds. Delays in initiating therapy, conversely, are eclipsed in importance when appraised in conjunction with the entire treatment course. Such findings may redirect focus to streamlining those intervals that are most sensitive to delays when considering survival burden. Cancer 2018. © 2018 American Cancer Society.

Predictors of adverse outcome in uterine smooth muscle tumours of uncertain malignant potential (STUMP): a clinicopathological analysis of 22 cases with a proposal for the inclusion of additional histological parameters.

AIMS: The current World Health Organisation classification defines smooth muscle tumours of uncertain malignant potential (STUMPs) as neoplasms that cannot be diagnosed reliably as benign or malignant according to generally accepted criteria. This has led to the application of various sets of criteria; consequently, consistent and reliable outcome data are lacking. The aims of this study were: (i) to compare the frequency of adverse outcome in STUMP on the basis of enhanced criteria; and (ii) to perform failure analysis to identify feature(s) helpful in predicting outcome METHODS AND RESULTS: Cases of STUMP diagnosed between 1994 and 2009 were retrieved and follow-up data were collected. Morphological parameters were scored and correlated with outcome. Twenty-two subjects with a median follow-up of 74.5 months (range, 26-166 months) formed the study group. Their age ranged from 31.9 years to 51.8 years (median, 45.3 years). Sixteen subjects underwent hysterectomy and six underwent myomectomy. Adverse outcomes were noted in eight (36.4%) cases. In cases with adverse outcomes, notable features included moderate-severe nuclear atypia (seven), epithelioid features (one), infiltrative or irregular margins (five), atypical mitoses (two), and vascular intrusion (three) CONCLUSIONS: The frequency of adverse outcomes in our series (36.4%) was higher than that in previously published reports (7-26.7%), suggesting that the use of more stringent criteria can exclude some patients from further follow-up. Although 'significant' nuclear atypia was not discriminatory, its frequent association with adverse outcomes has pathobiological implications. The presence of necrosis was not particularly associated with adverse outcomes. Atypical mitoses, epithelioid differentiation, vascular involvement and infiltrative/irregular margins appear to herald adverse outcomes, and therefore merit inclusion in the diagnostic regimen.

Challenges in Head and Neck Pathology.

The surgical pathologist plays a crucial role in the multidisciplinary team. In nearly all cases, a tissue diagnosis is required to confirm the disease process before treatment begins. Even in settings where the diagnosis appears straightforward, a timely and appropriate report is necessary. The pathologist is also responsible for providing many of the more specific data elements that will guide treatment decisions: examples include evidence of virally driven malignancy, margin status, and the precise depth to which tumor invades. Each of these diagnoses and findings has its own specific set of difficulties and limitations, which require nuanced interpretation by a well-informed pathologist.

Association of Quantitative Metastatic Lymph Node Burden With Survival in Hypopharyngeal and Laryngeal Cancer.

Importance: Nodal staging for laryngohypopharyngeal cancers is based primarily on size and laterality, with less value placed on absolute number of metastatic lymph nodes (LNs). We are aware of no studies to date that have specifically addressed the prognostic effect of quantitative nodal burden in larynx or hypopharynx malignancies. Objective: To assess the independent impact of quantitative metastatic LN burden on mortality risk. Design, Setting, and Participants: Univariate and multivariable models were constructed to evaluate the association between patients' number of metastatic LNs and their survival, adjusting for factors such as nodal size, laterality, extranodal extension, margin status, and adjuvant treatment. Participants were patients with squamous cell carcinoma of the larynx or hypopharynx undergoing upfront surgical resection for curative intent at a US hospital between 2004 and 2013, as identified in the National Cancer Database. A neck dissection of a minimum of 10 LNs was required. Main Outcomes and Measures: Overall survival. Results: Overall, 8351 cases were included (mean [SD] age, 61 [10.1] years; 6499 men [77.8%]; 4710 patients with metastatic LNs and 3641 with no metastatic LNs). Mortality risk escalated continuously without plateau as number of metastatic nodes increased, with the hazard per node (hazard ratio [HR], 1.19; 95% CI, 1.16-1.23; P < .001) most pronounced up to 5 positive LNs. Extranodal extension was also associated with increased mortality (HR, 1.34; 95% CI, 1.13-1.59; P < .001). Increasing number of nodes examined was associated with improved survival, albeit to a lesser degree (per 10 LNs: HR, 0.97; 95% CI, 0.96-0.98; P < .001) and without a detectable change point. Other nodal factors, including nodal size, contralateral LN involvement (TNM stage N2c), and lower LN involvement (levels 4-5), were not associated with mortality in multivariable models when accounting for number of positive LNs. A novel, parsimonious nodal staging system derived by recursive partitioning analysis exhibited greater concordance with survival than the TNM staging system outlined in the American Joint Committee on Cancer's AJCC Staging Manual, 8th edition. Conclusions and Relevance: The number of metastatic nodes is a predominant independent factor associated with mortality in hypopharyngeal and laryngeal cancers. Moreover, standard nodal staging factors like LN size and contralaterality have no independent prognostic value when accounting for positive LN number. Deeper integration of quantitative metastatic nodal disease may simplify staging and better triage the need for adjuvant therapy.

Metastatic Lymph Node Burden and Survival in Oral Cavity Cancer.

Purpose Current staging systems for oral cavity cancers incorporate lymph node (LN) size and laterality, but place less weight on the total number of positive metastatic nodes. We investigated the independent impact of numerical metastatic LN burden on survival. Methods Adult patients with oral cavity squamous cell carcinoma undergoing upfront surgical resection for curative intent were identified in the National Cancer Data Base between 2004 and 2013. A neck dissection of a minimum of 10 LNs was required. Multivariable models were constructed to assess the association between the number of metastatic LNs and survival, adjusting for factors such as nodal size, laterality, extranodal extension, margin status, and adjuvant treatment. Results Overall, 14,554 patients met inclusion criteria (7,906 N0 patients; 6,648 node-positive patients). Mortality risk escalated continuously with increasing number of metastatic nodes without plateau, with the effect most pronounced with up to four LNs (HR, 1.34; 95% CI, 1.29 to 1.39; P < .001). Extranodal extension (HR, 1.41; 95% CI, 1.20 to 1.65; P < .001) and lower neck involvement (HR, 1.16; 95% CI, 1.06 to 1.27; P < .001) also predicted increased mortality. Increasing number of nodes examined was associated with improved survival, plateauing at 35 LNs (HR, 0.98; 95% CI, 0.98 to 0.99; P < .001). In multivariable models accounting for the number of metastatic nodes, contralateral LN involvement (N2c status) and LN size were not associated with mortality. A novel nodal staging system derived by recursive partitioning analysis exhibited greater concordance than the American Joint Committee on Cancer (8th edition) system. Conclusion The number of metastatic nodes is a critical predictor of oral cavity cancer mortality, eclipsing other features such as LN size and contralaterality in prognostic value. More robust incorporation of numerical metastatic LN burden may augment staging and better inform adjuvant treatment decisions.

Directed Differentiation of Human Induced Pluripotent Stem Cells into Fallopian Tube Epithelium.

The fallopian tube epithelium (FTE) has been recognized as a site of origin of high-grade serous ovarian cancer (HGSC). However, the absence of relevant in vitro human models that can recapitulate tissue-specific architecture has hindered our understanding of FTE transformation and initiation of HGSC. Here, induced pluripotent stem cells (iPSCs) were used to establish a novel 3-dimensional (3D) human FTE organoid in vitro model containing the relevant cell types of the human fallopian tube as well as a luminal architecture that closely reflects the organization of fallopian tissues in vivo. Modulation of Wnt and BMP signaling directed iPSC differentiation into Müllerian cells and subsequent use of pro-Müllerian growth factors promoted FTE precursors. The expression and localization of Müllerian markers verified correct cellular differentiation. An innovative 3D growth platform, which enabled the FTE organoid to self-organize into a convoluted luminal structure, permitted matured differentiation to a FTE lineage. This powerful human-derived FTE organoid model can be used to study the earliest stages of HGSC development and to identify novel and specific biomarkers of early fallopian tube epithelial cell transformation.

Pattern-based classification of invasive endocervical adenocarcinoma, depth of invasion measurement and distinction from adenocarcinoma in situ: interobserver variation among gynecologic pathologists.

A pattern-based classification for invasive endocervical adenocarcinoma has been proposed as predictive of the risk of nodal metastases. We aimed to determine the reproducibility of such classification in the context of common diagnostic challenges: distinction between in situ and invasive adenocarcinoma and depth of invasion measurement. Nine gynecologic pathologists independently reviewed 96 cases of endocervical adenocarcinoma (two slides per case). They diagnosed each case as in situ or invasive carcinoma classifying the latter following the pattern-based classification as pattern A (non-destructive), B (focally destructive) or C (diffusely destructive). Depth of invasion, when applicable, was measured (mm). Overall, diagnostic reproducibility of pattern diagnosis was good (κ=0.65). Perfect agreement (9/9 reviewers) was seen in only 11 cases (11%), all destructively invasive (10 pattern C and 1 pattern B). In all, ≥5/9 reviewer concordance was achieved in 82/96 cases (85%). Distinction between in situ and invasive carcinoma, regardless of the pattern, showed only slight agreement (κ=0.37). Likewise, distinction restricted to in situ versus pattern A was poor (κ=0.23). Distinction between non-destructive (in situ+pattern A) and destructive (patterns B+C) carcinoma showed significantly higher agreement (κ=0.62). Estimation of depth of invasion showed excellent reproducibility (ICC=0.82). However, different measurements resulting in differing FIGO stages were common (from at least 1 reviewer in 79% cases). On the basis of interobserver agreement, the pattern-based classification is best at diagnosing destructive invasion, which carries a risk for nodal metastases. Agreement in diagnosing in situ versus invasive carcinoma, including pattern A, was poor. Given the nil risk of nodal spread in in situ and pattern A lesions, the term 'endocervical adenocarcinoma with non-destructive growth' can be considered when the distinction is difficult, after excluding destructive invasion. Depth of invasion measurement was highly reproducible among pathologists; thus, the pattern-based approach can complement, but should not replace, the depth of invasion metric.

Comparison of metastatic neuroendocrine neoplasms to the breast and primary invasive mammary carcinomas with neuroendocrine differentiation.

Metastatic neuroendocrine neoplasms to the breast may show considerable morphologic overlap with primary mammary carcinomas, particularly those showing evidence of neuroendocrine differentiation, and may be misdiagnosed as such. Accurate distinction between these two entities is crucial for determination of appropriate clinical management. The histologic and immunohistochemical features of metastatic neuroendocrine neoplasms to the breast were studied and compared with the features of primary invasive mammary carcinomas with neuroendocrine differentiation, which served as controls. Of the metastatic neuroendocrine neoplasms, 15 were well-differentiated neuroendocrine tumors with carcinoid tumor-type morphology and 7 were poorly differentiated/high-grade neuroendocrine carcinomas with small-cell or large-cell neuroendocrine carcinoma morphology. The majority of the metastatic neoplasms originated in the lung and gastrointestinal tract. There were histologic similarities between metastatic neuroendocrine neoplasms and invasive mammary carcinomas with neuroendocrine differentiation, both of which exhibited neuroendocrine histologic features (nested and trabecular architecture, minimal tubular differentiation, and characteristic nuclear features). Only one case of the invasive mammary carcinomas with neuroendocrine differentiation was modified Bloom-Richardson grade 1 (largely due to minimal tubular differentiation on most such tumors), and the invasive mammary carcinomas with neuroendocrine differentiation were often associated with in situ carcinoma. Immunohistochemistry was helpful in distinguishing metastatic neuroendocrine neoplasms from invasive mammary carcinomas with neuroendocrine differentiation. Whereas the majority of invasive mammary carcinomas with neuroendocrine differentiation were positive for estrogen receptor and GATA3, metastatic neuroendocrine neoplasms were typically negative for estrogen receptor and GATA3, and metastatic well-differentiated neuroendocrine tumors often showed immunoreactivity for site-specific markers. Although the histologic and immunohistochemical features of a breast tumor may raise the suspicion of a metastatic neuroendocrine neoplasm, the pathologic findings should be interpreted in the context of the clinical history and imaging findings in order to establish an accurate diagnosis.

Cervical squamocolumnar junction-specific markers define distinct, clinically relevant subsets of low-grade squamous intraepithelial lesions.

Low-grade cervical squamous abnormalities (low-grade squamous intraepithelial lesions [LSIL, CIN1]) can be confused with or followed by high-grade (HSIL, CIN2/3) lesions, expending considerable resources. Recently, a cell of origin for cervical neoplasia was proposed in the squamocolumnar junction (SCJ); HSILs are almost always SCJ, but LSILs include SCJ and SCJ subsets. Abnormal cervical biopsies from 214 patients were classified by 2 experienced pathologists (panel) as LSIL or HSIL using published criteria. SILs were scored SCJ and SCJ using SCJ-specific antibodies (keratin7, AGR2, MMP7, and GDA). Assessments of interobserver agreement, p16 staining pattern, proliferative index, and outcome were compared. The original diagnostician agreed with the panel diagnosis of HSIL and SCJ LSIL in all cases (100%). However, for SCJ LSIL, panelists disagreed with each other by 15% and with the original diagnostician by 46.2%. Comparing SCJ and SCJ LSILs, 60.2% and 94.9% were p16 positive, 23% and 74.4% showed strong (full-thickness) p16 staining, and 0/54 (0%) and 8/33 (24.2%) with follow-up had an HSIL outcome, respectively. Some SCJ LSILs are more likely to both generate diagnostic disagreement and be associated with HSIL. Conversely, SCJ LSILs generate little observer disagreement and, when followed, have a very low risk of HSIL outcome. Thus, SCJ biomarkers in conjunction with histology may segregate LSILs with very low risk of HSIL outcome and conceivably could be used as a management tool to reduce excess allocation of resources to the follow-up of these lesions.

Oncogenic mutations in cervical cancer: genomic differences between adenocarcinomas and squamous cell carcinomas of the cervix.

BACKGROUND: Cervical cancer is the second leading cause of cancer deaths among women worldwide. The objective of this study was to describe the most common oncogenic mutations in cervical cancers and to explore genomic differences between the 2 most common histologic subtypes: adenocarcinoma and squamous cell carcinoma. METHODS: A high-throughput genotyping platform, termed Oncomap, was used to interrogate 80 cervical tumors for 1250 known mutations in 139 cancer genes. Samples were analyzed using a mass spectrometry-based genotyping platform and were validated using orthogonal chemistry. Epidermal growth factor receptor (EGFR) mutations were further validated by massive parallel sequencing. Human papilloma virus (HPV) genotyping also was performed. RESULTS: Validated mutations were detected in 48 of 80 tumors (60%) examined. The highest mutation rates were in the genes phosphatidylinositol 3-kinase, catalytic subunit α (PIK3CA) (31.3%); Kirsten rat sarcoma viral oncogene homolog (KRAS) (8.8%); and EGFR (3.8%). PIK3CA mutation rates did not differ significantly between adenocarcinomas and squamous cell carcinomas (25% vs 37.5%, respectively; P = .33). In contrast, KRAS mutations were identified only in adenocarcinomas (17.5% vs 0%; P = .01), and a novel EGFR mutation was detected only in squamous cell carcinomas (0% vs 7.5%; P = .24). There were no associations between HPV-16 or HPV-18 and somatic mutations or overall survival. In adjusted analyses, PIK3CA mutations were associated with shorter survival (67.1 months vs 90.3 months; hazard ratio, 9.1; 95% confidence interval, 2.8-29.5 months; P < .001). CONCLUSIONS: Cervical cancers harbor high rates of potentially targetable oncogenic mutations. In addition, cervical squamous cell carcinoma and adenocarcinoma have distinct molecular profiles, suggesting that clinical outcomes may be improved with the use of more tailored treatment strategies, including PI3K and MEK inhibitors.

Aspergillus fumigatus vegetation of a prosthetic aortic root graft with mycotic aneurysm and subarachnoid hemorrhage.

A 58-year-old woman with a history of Bentall aortic graft and bioprosthetic aortic valve replacement 3 months prior to admission, presented with headache and fever. Imaging yielded a large obstructive filling defect in the ascending aorta, a subarachnoid hemorrhage, and a mycotic aneurysm. Intraoperative specimens grew Aspergillus fumigatus, and despite aggressive measures the patient died. Aspergillus infections of prosthetic vascular grafts are rare surgical complications and are difficult to diagnose given the low incidence of positive microbiology cultures and the long median time between surgery and diagnosis. Treatment has consisted of antifungal and surgical treatment, although mortality remains high.

Utility of PAX2 as a marker for diagnosis of endometrial intraepithelial neoplasia.

Diagnosis of endometrial intraepithelial neoplasia (EIN) requires learning new criteria. Two trainees rendered diagnoses based on biopsy findings, and then measured the effect of reviewing PAX2 on their interpretation. Fifty-two endometrial biopsy specimens diagnosed as having EIN were evaluated using EIN criteria. Background endometrial pattern, altered differentiation, and any features complicating diagnosis were noted. PAX2 stains were scored as confusing, helpful, or noncontributory. Fifty-two cases generated 104 passes; 82% were rediagnosed as EIN. The diagnosis was complicated because of altered differentiation (14%), EIN and background separation (13%), large lesions lacking background (11%), and secretory background (8%). PAX2 was most helpful in cases with secretory backgrounds and when EIN lacked adjacent normal tissue, and most confusing when scoring was ambiguous (14%). The diagnosis of EIN can be difficult when: (1) the lesion cannot be easily compared with background; (2) there is a confounding process; and (3) gland differentiation is altered. PAX2 can be of assistance in delimiting EIN lesions.