Performance of HER2 DAKO HercepTest and Ventana 4B5 immunohistochemical assays on detecting HER2 gene-amplification in uterine serous carcinomas

We compared the performance of two commonly-used HER2 immunohistochemistry (IHC) assays in uterine serous carcinomas (USC), correlating with HER2 gene amplification by fluorescence in-situ hybridization (FISH). Sixty-five USCs were stained by both HercepTest™ and PATHWAY 4B5 assays. FISH was performed by HER2 IQFISH pharmDx. Consensus HER2 IHC scoring was performed, and HER2 testing results were evaluated using USC-specific criteria. Complete concordance between HercepTest and 4B5 assays was achieved in 44/65 tumors (68%). The overall HER2 IHC/FISH concordance was 94% (45/48) by HercepTest and 91% (42/46) by 4B5. All HER2 IHC 3+ cases with HercepTest (n = 6) and 4B5 (n = 4) were gene-amplified, corresponding to specificities of 100%. For cases with IHC 2+, 41% (7/17) by HercepTest and 42% (8/19) by 4B5 had HER2 gene amplification. The sensitivity for HercepTest and 4B5 were 38% and 25%, respectively, at a cut-off of IHC 3+ (P = 0.50), and were 81% and 75%, respectively, at a cut-off of IHC 2+ (P > 0.99). Among HER2 IHC 0-1+ cases, 3/42 cases by HercepTest and 4/42 cases by 4B5 showed amplified FISH results, corresponding to overall false negative rates of 19% for HercepTest and 25% for 4B5. By using USC-specific IHC scoring criteria, both HercepTest and 4B5 assays showed high specificities (100%) for HER2 gene amplification in IHC 3+ cases, high IHC/FISH concordance, and comparable sensitivity for detecting HER2 gene amplification. The notable false negative rates using IHC 2+ as a cut-off for reflexing FISH analysis may warrant consideration for performing FISH in IHC 1+ cases until more data become available.

HER2 evaluation for clinical decision making in human solid tumours: pearls and pitfalls

The significant clinical benefits of human epidermal growth factor receptor 2 (HER2)-targeted therapeutic agents have revolutionized the clinical treatment landscape in a variety of human solid tumours. Accordingly, accurate evaluation of HER2 status in these different tumour types is critical for clinical decision making to select appropriate patients who may benefit from life-saving HER2-targeted therapies. HER2 biomarker scoring criteria is different in different organ systems, and close adherence to the corresponding HER2 biomarker testing guidelines and their updates, if available, is essential for accurate evaluation. In addition, knowing the unusual patterns of HER2 expression is also important to avoid inaccurate evaluation. In this review, we discuss the key considerations when evaluating HER2 status in solid tumours for clinical decision making, including tissue handling and preparation for HER2 biomarker testing, as well as pathologist's readout of HER2 testing results in breast carcinomas, gastroesophageal adenocarcinomas, colorectal adenocarcinomas, gynaecologic carcinomas, and non-small cell lung carcinomas.

In ER-Positive, HER2-Negative Breast Cancers, HER2 mRNA Levels Correlate Better with Clinicopathologic Features and Oncotype DX Recurrence Score than HER2 Immunohistochemistry

With the approval of trastuzumab deruxtecan for treating advanced human epidermal growth factor receptor-2 (HER2) low breast cancer (BC), it has become increasingly important to develop more accurate and reliable methods to identify HER2-low BC. In addition, HER2 immunohistochemistry (IHC) has limitations for quantification of HER2. We explored the relationship between HER2 IHC and mRNA levels and evaluated whether HER2 IHC scores and mRNA levels are associated with clinicopathologic features and Oncotype DX Recurrence Score (RS) in estrogen receptor (ER)-positive, HER2-negative BCs. A total of 750 BCs sent for Oncotype DX (ODX) testing were included in this study, and 559 with HER2 mRNA levels were available. There were no statistically significant differences between HER2 0 and HER2-low BC in clinicopathologic variables or ODX RS using HER2 IHC. There was a significant difference in median HER2 mRNA values between HER2 0 and HER2-low (8.7 vs 9.3, P < .001); however, the HER2 mRNA distribution had substantial overlap between these 2 groups with a suboptimal area under the receiver operating characteristic curve (area under the receiver operating characteristic curve = 0.68). A HER2 mRNA value of 9.2 was generated as the optimal cutoff for distinguishing HER2 0 and HER2-low BC. Comparing ER+ BCs with HER2 mRNA high (>9.2) and low (≤9.2) revealed a statistically significant difference in most clinicopathologic variables and ODX RS. From this large cohort of ER-positive, HER2-negative BC, our results demonstrated that HER2 mRNA levels correlated better with clinicopathologic features and recurrence risk as assessed by ODX RS than HER2 IHC scores. Our findings suggest that HER2 mRNA-detecting methods could potentially serve as a quantitative and reliable method for identifying a biologically meaningful group of HER2-low BC. Further study is needed to determine whether HER2 mRNA levels could be more reliable than IHC for identifying which patients will be most likely to benefit from trastuzumab deruxtecan.

HER2 in uterine serous carcinoma: Current state and clinical perspectives

OBJECTIVES

Uterine cancer has the highest incidence and the second-highest mortality rate among gynecologic malignancies in the United States. Although uterine serous carcinoma (USC) represents less than 10% of endometrial carcinomas, it accounts for a disproportionate 50% of tumor relapses and 40% of endometrial cancer deaths. Over the past decade, clinical trials have focused on finding better treatments for this aggressive subtype of endometrial cancer, especially HER2-targeted therapy.

METHODS

We conducted a literature search in PubMed to expand the understanding of HER2 in USC.

RESULTS

HER2 has been established as an important biomarker with prognostic and therapeutic implications in USC. Intratumoral heterogeneity and lateral/basolateral membranous staining of HER2 as well as high discordance between HER2 immunohistochemistry and in situ hybridization are more common in USC than in breast carcinoma. Therefore, a universal HER2 testing and scoring system more suitable to endometrial cancer is needed and currently under investigation.

CONCLUSIONS

This review discusses the clinical perspective of HER2 overexpression/gene amplification in USC, the distinct HER2 staining pattern and the evaluation of HER2 in USC, the resistance mechanisms of HER2-targeted therapy in HER2-positive cancers, and likely areas of future investigation.

Ovarian Combined Serous Borderline Tumor/Low-grade Serous Carcinoma and Mesonephric-like Lesion: Report of 2 Cases With New Observations

Ovarian combined serous borderline tumor/low-grade serous carcinomas (SBT/LGSC) and mesonephric-like adenocarcinomas (MLA) have been previously reported and the presence of identical oncogenic somatic mutations in both components supports the concept that at least some of MLAs arise from a Müllerian origin. We report 2 cases of ovarian combined SBT/LGSC and mesonephric-like lesion. Case 1 was a 70-yr-old woman presented with a liver lesion and omental carcinomatosis. Histologic examination revealed biphasic tumors in bilateral ovaries consisting of conventional SBT and invasive MLA with extraovarian spread. The right ovary also had a component of cribriform variant of SBT/noninvasive LGSC. The SBT/LGSC component was diffusely positive for Pax8, WT-1, and ER, focally positive for PR, and negative for GATA3, while the MLA component was diffusely positive for GATA3 but negative for WT-1, ER, and PR. Molecular analysis revealed a KRAS G12V mutation in both the SBT/LGSC and MLA components, indicating their clonal origin. Case 2 was a 58-yr-old woman who presented with conventional type SBT in both ovaries. In addition, the left ovarian tumor demonstrated a few areas (each <5 mm) of mesonephric-like differentiation/hyperplasia in close proximity to the serous-type epithelium, with an immunophenotype of focal GATA3 expression, luminal pattern of CD10 staining and negative WT-1, ER, and PR staining. This phenomenon has been reported in endometrioid borderline tumor but not in any serous type lesions. The findings in case 1 provide further evidence to demonstrate the clonal relationship between these morphologically and immunophenotypically distinct components. It also supports the theory that, unlike cervical mesonephric carcinomas originating from mesonephric remnants, MLAs are derived from a Müllerian-type lesion with differentiation into mesonephric lineage. The presence of a hyperplastic mesonephric-like lesion/differentiation in case 2 indicates that a precursor lesion in the same lineage with the potential to develop into MLA exists in the ovary.

The Rochester Modified Magee Algorithm (RoMMa): An Outcomes Based Strategy for Clinical Risk-Assessment and Risk-Stratification in ER Positive, HER2 Negative Breast Cancer Patients Being Considered for Oncotype DX® Testing

INTRODUCTION

Multigene genomic profiling has become the standard of care in the clinical risk-assessment and risk-stratification of ER⁺, HER2^- breast cancer (BC) patients, with Oncotype DX^® (ODX) emerging as the genomic profile test with the most support from the international community. The current state of the health care economy demands that cost-efficiency and access to testing must be considered when evaluating the clinical utility of multigene profile tests such as ODX. Several studies have suggested that certain lower risk patients can be identified more cost-efficiently than simply reflexing all ER⁺, HER2^- BC patients to ODX testing. The Magee equations^TM use standard histopathologic data in a set of multivariable models to estimate the ODX recurrence score. Our group published the first outcome data in 2019 on the Magee equations^TM, using a modification of the Magee equations^TM combined with an algorithmic approach-the Rochester Modified Magee algorithm (RoMMa). There has since been limited published outcome data on the Magee equations^TM. We present additional outcome data, with considerations of the TAILORx risk-stratification recommendations.

METHODS

355 patients with an ODX recurrence score, and at least five years of follow-up or a BC recurrence were included in the study. All patients received either Tamoxifen or an aromatase inhibitor. None of the patients received adjuvant systemic chemotherapy.

RESULTS

There was no significant difference in the risk of recurrence in similar risk categories (very low risk, low risk, and high risk) between the average Modified Magee score and ODX recurrence score with the chi-square test of independence (p > 0.05) or log-rank test (p > 0.05). Using the RoMMa, we estimate that at least 17% of individuals can safely avoid ODX testing.

CONCLUSION

Our study further reinforces that BC patients can be confidently stratified into lower and higher-risk recurrence groups using the Magee equations^TM. The RoMMa can be helpful in the initial clinical risk-assessment and risk-stratification of BC patients, providing increased opportunities for cost savings in the health care system, and for clinical risk-assessment and risk-stratification in less-developed geographies where multigene testing might not be available.

The Evolution of Ki-67 and Breast Carcinoma: Past Observations, Present Directions, and Future Considerations

The 1983 discovery of a mouse monoclonal antibody-the Ki-67 antibody-that recognized a nuclear antigen present only in proliferating cells represented a seminal discovery for the pathologic assessment of cellular proliferation in breast cancer and other solid tumors. Cellular proliferation is a central determinant of prognosis and response to cytotoxic chemotherapy in patients with breast cancer, and since the discovery of the Ki-67 antibody, Ki-67 has evolved as an important biomarker with both prognostic and predictive potential in breast cancer. Although there is universal recognition among the international guideline recommendations of the value of Ki-67 in breast cancer, recommendations for the actual use of Ki-67 assays in the prognostic and predictive evaluation of breast cancer remain mixed, primarily due to the lack of assay standardization and inconsistent inter-observer and inter-laboratory reproducibility. The treatment of high-risk ER-positive/human epidermal growth factor receptor-2 (HER2) negative breast cancer with the recently FDA-approved drug abemaciclib relies on a quantitative assessment of Ki-67 expression in the treatment decision algorithm. This further reinforces the urgent need for standardization of Ki-67 antibody selection and staining interpretation, which will hopefully lead to multidisciplinary consensus on the use of Ki-67 as a prognostic and predictive marker in breast cancer. The goals of this review are to highlight the historical evolution of Ki-67 in breast cancer, summarize the present literature on Ki-67 in breast cancer, and discuss the evolving literature on the use of Ki-67 as a companion diagnostic biomarker in breast cancer, with consideration for the necessary changes required across pathology practices to help increase the reliability and widespread adoption of Ki-67 as a prognostic and predictive marker for breast cancer in clinical practice.

The role of PRAME and NY-ESO-1 as potential therapeutic and prognostic biomarkers in triple-negative breast carcinomas

PRAME and NY-ESO-1 are cancer-testis antigens (CTAs) reported to be highly enriched in triple-negative breast cancers (TNBCs), against which vaccines and immunotherapies are currently being developed. This study aims to analyze PRAME and NY-ESO-1 expression in TNBCs and their correlation with clinical outcomes. This is a retrospective cohort study of TNBC patients who have undergone neoadjuvant chemotherapy. PRAME and NY-ESO-1 expression were assessed on pre-therapy biopsies as H-scores (percentage x intensity) with final H scores of 2-3 considered as positive. Association between expression and pathologic complete response (pCR), metastasis, and residual cancer burden (RCB) were assessed via logistic regression. Cox proportional hazards models were used to assess the association with progression-free survival. P-values < 0.05 were considered statistically significant. Sixty-three percent of 76 patients were positive for PRAME. In contrast, only 5 % were positive for NY-ESO-1. PRAME positivity was significantly associated with a lower likelihood of early metastatic disease (OR = 0.24, 95 % CI 0.08-0.62; P = 0.005). However, it was not significantly associated with pCR, RCB category, or progression-free survival. NY-ESO1 score was not significantly associated with early metastatic disease, pCR, RCB category, or progression-free survival. Our results suggest that PRAME positivity may be associated with a lower risk of early metastasis in TNBCs, but not with response to neoadjuvant chemotherapy or progression-free survival. The high expression of PRAME in TNBCs makes it a potential therapeutic target, while NY-ESO1 appears to be a less useful marker. However, further larger studies are needed to ascertain the utility of these markers.

Mucinous Borderline Tumor Associated with Mesonephric-like Proliferation: Further Evidence for a Possible New Origin of Ovarian Mucinous Neoplasms

Some ovarian mucinous tumors are thought to originate from Brenner tumors and teratomas; however, data are limited on what could be the origin for the remaining tumors. We report a new case of ovarian mucinous borderline tumor/atypical proliferative mucinous tumor (MBT/APMT) co-existing with a mesonephric-like proliferation (MLP)/mesonephric-like hyperplasia (MLH). The patient was a 58-year-old woman who presented with a pelvic mass and abdominal pain. Pathology demonstrated an 11 cm MBT/APMT in the left ovary. In addition, the tumor contained one focal area (<1% of total tumor volume) of MLP/hyperplasia adjacent to, or intimately admixed with, mucinous epithelium, with an immunophenotype of diffuse Pax8 and Gata3 expression and negative TTF-1, ER, and PR staining. Pax8 was also weakly positive in the MBT/APMT component. Some mesonephric-like glands partially exhibited gastrointestinal-type mucinous metaplasia/differentiation. A polymerase chain reaction (PCR)-based Sanger sequencing demonstrated that a KRAS G12V mutation was present in both MLP/MLH and MBT/APMT components, providing further evidence to support their clonal relationship. We previously reported a series of similar cases and demonstrated a novel association between MLP, mesonephric-like adenocarcinoma and ovarian mucinous tumor. It is conceivable that benign MLPs may have ability to differentiate to lineage-specific mucinous lesions, and, as such, they may serve as a possible new origin of some ovarian mucinous neoplasms; in particular, Pax8-positive tumors. The current case provides additional evidence to support this theory.

Mucinous Tumor Coexisting With Mesonephric-like Proliferation/Tumor in the Ovary: A Novel Association

The literature indicates that mesonephric carcinoma (MC) and mesonephric-like adenocarcinoma (MLA) typically lack mucinous and squamous features/differentiation. We report 4 cases of ovarian mucinous tumors (1 mucinous cystadenofibroma and 3 mucinous borderline tumors/atypical proliferative mucinous tumors [MBT/APMT]) co-existing with mesonephric-like lesions which were highlighted by Gata3 and Pax8 expression. All cases contained benign mesonephric-like proliferations (MLP) which focally displayed gastrointestinal-type mucinous metaplasia/differentiation and some were intimately admixed with mucinous glands associated with the mucinous tumor. Metaplastic mucinous epithelium retained expression of Gata3 and Pax8 in some areas while 1 mucinous cystadenofibroma and 1 MBT/APMT were focally positive for Pax8. Along with these mesonephric components, case 1 exhibited features of mesonephric hyperplasia and in 2 cases, 3 and 4, MLA was identified. In case 4, a KRAS c.35G>T (p.Gly12Val) somatic mutation was detected in both the MBT/APMT and the MLA, indicating a clonal origin. This same mutation was also detected in the benign MLP, indicating that it was likely an early genetic event. A CTNNB1 c.98C>T (p.Ser33Phe) somatic mutation, FGFR2 amplification, and CDKN2A/p16 deletion were only detected in the MLA but not in the MBT/APMT. Our result provides evidence to demonstrate the clonal relationship between these morphologically distinct components. Although speculative, we postulate that benign MLPs may give rise to lineage-specific mucinous and mesonephric-like lesions and propose that the MLPs are a new possible origin of some ovarian mucinous tumors. Whether these MLPs arise through transdifferentiation of Müllerian tissue or represent true mesonephric remnants, however, remains largely unknown.

Over-expression of Programmed Death Ligand 1 (PD-L1) in Refractory Inflammatory Bowel Disease (IBD)

Programmed death ligand 1 (PD-L1) dysregulation has been implicated in chronic inflammatory diseases, but its role in regulating intestinal mucosa inflammation is still unclear. The aim of this study was to assess PD-L1 expression in the intestinal mucosa of patients with refractory inflammatory bowel disease (IBD) compared to controls. We evaluated PD-L1 expression by immunohistochemistry in colectomy specimens of patients with ulcerative colitis (UC) and Crohn disease (CD) compared to controls. PD-L1 expression was assessed in colonic epithelium and inflammatory cells, along with the location of the inflammatory cells expressing PD-L1. All cases were stained with CD3, CD4, CD8, FOXP3, CD20, CD68, and CD90 immunostains to determine the types of cells expressing PD-L1. The UC group showed significantly higher PD-L1 expression in the colonic epithelium compared to both CD and control groups (both P<0.001), and CD was also significantly higher than the control group (P=0.004). Both UC and CD groups showed similar PD-L1 expression in the inflammatory infiltrate, but significantly higher than the control group (both P<0.001). Among both IBD groups, higher IBD activity was associated with higher levels of PD-L1 expression in the colonic epithelium (P<0.05) and inflammatory infiltrate (P<0.001). When comparing PD-L1 expression to lineage specific markers, CD3+, CD4+ T cells, CD68+ macrophages, and CD90+ colonic stromal cells appeared to be expressing PD-L1. These findings implicate a role for PD-L1 in the dysregulation of the immune response in refractory IBD. Further studies are warranted to better understand the role of the immune regulatory pathways in intestinal mucosa.

Abstract P1-02-11: Somatic alterations and PD-L1 positivity in advanced breast cancer

Background: With the recent FDA approvals of atezolizumab and pembrolizumab for the treatment of unresectable locally advanced or metastatic triple-negative breast cancer, immuno-oncology (IO) therapy has become feasible for the first time in breast cancer, and PD-L1 immunohistochemistry (IHC) has become an essential assay in breast cancer diagnostics. However, relatively little is known about the molecular profile of PD-L1+ breast cancers compared to other tumors, such as lung cancer. Furthermore, although the initial data have shown the most benefit in triple-negative patients, it is unclear whether additional subgroups of patients with advanced breast cancer enriched for PD-L1 expression can be identified, and, therefore, might benefit from IO therapy. Methods: We identified 60 women with locally advanced or metastatic breast cancer who underwent next generation sequencing (NGS) of tumor samples obtained at our institution as part of their clinical care and performed PD-L1 testing using the Ventana SP142 platform. Samples were considered positive for PD-L1 if ≥1%. The NGS assay was designed to detect single nucleotide variants, insertions and deletions, copy number variants, and specific translocations in approximately 600 genes. Associations between PD-L1 status, molecular alterations, and tumor histopathologic features were assessed, with P &lt; 0.05 considered statistically significant. Results: Our cohort consisted of 60 women at a median age of 56 years (range 31-81). 50% of samples were from primary tumors within the breast, 15% were from regional lymph node metastases, and 35% were from distant metastatic sites. Most tumors showed ductal histology (80%) and were either grade 2 (38%) or grade 3 (57%), while half showed focal (27%) or extensive (23%) necrosis. Based on IHC, most tumors were Luminal B (54%), followed by triple-negative (25%), Luminal A (13%), and HER2-enriched (8%) tumors. About one third of tumors had either moderate (25%) or marked (8%) TILs, and about half showed positivity for PD-L1 (48%). Most tumors had a low tumor mutational burden (TMB), with 83% having &lt;5, 14% having 5-9, and 3% having ≥10 mutations per Mb. The median number of genes with reportable somatic alterations was 2 (IQR 1-3). The most common genes with somatic alterations were TP53 (51%), PIK3CA (23%), ERBB2 (16%), ESR1 (14%), FGFR1 (9%), AKT1 (7%), CDH1 (7%), GATA3 (7%), MYC (7%), and CCND1 (5%). 50% of luminal A, 38% of luminal B, 50% of HER2-enriched, and 75% of triple-negative tumors were PD-L1+, although this difference was not statistically significant (P = 0.22). PD-L1 positivity was more likely in grade 3 tumors (62% vs 31% grade 1-2, P = 0.02), and in tumors with extensive necrosis (85% vs 37% absent/focal, P = 0.004). Of all the genes tested, only TP53 was significantly associated with being PD-L1+ (69% TP53-mutant vs 32% TP53-wild-type; P = 0.008). Neither number of somatic alterations (P = 0.60) or TMB (P = 0.08) were significantly associated with PD-L1+ tumors. Of note, 53% of non-triple-negative PD-L1+ tumors had mutations in TP53, and 60% of non-triple-negative tumors with TP53 mutations were PD-L1+. Conclusions: Our data suggest that PD-L1 positivity is relatively common in patients with locally advanced or metastatic breast cancer, not only in triple-negative breast cancer but in all breast cancer subtypes. In addition, TP53 mutation is significantly associated with an increased likelihood of positivity for PD-L1, including in non-triple-negative tumors. This correlation could potentially be a useful marker for selecting non-triple-negative breast cancer patients for PD-L1 testing and, if positive, potentially offer them IO options for optimizing their response to therapy.

Citation Format: Brian S Finkelman, Massimo Cristofanilli, Luis Z Blanco, Jr, Amir Behdad, Leonidas C Platanias, William J Gradishar, Kalliopi P Siziopikou. Somatic alterations and PD-L1 positivity in advanced breast cancer [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P1-02-11.

Correlation of manual semi-quantitative and automated quantitative Ki-67 proliferative index with OncotypeDXTM recurrence score in invasive breast carcinoma

BACKGROUND

Ki-67 immunohistochemistry (IHC) staining is a widely used cancer proliferation assay; however, its limitations could be improved with automated scoring. The OncotypeDXTM Recurrence Score (ORS), which primarily evaluates cancer proliferation genes, is a prognostic indicator for breast cancer chemotherapy response; however, it is more expensive and slower than Ki-67.

OBJECTIVE

To compare manual Ki-67 (mKi-67) with automated Ki-67 (aKi-67) algorithm results based on manually selected Ki-67 "hot spots" in breast cancer, and correlate both with ORS.

METHODS

105 invasive breast carcinoma cases from 100 patients at our institution (2011-2013) with available ORS were evaluated. Concordance was assessed via Cohen's Kappa (κ).

RESULTS

57/105 cases showed agreement between mKi-67 and aKi-67 (κ 0.31, 95% CI 0.18-0.45), with 41 cases overestimated by aKi-67. Concordance was higher when estimated on the same image (κ 0.53, 95% CI 0.37-0.69). Concordance between mKi-67 score and ORS was fair (κ 0.27, 95% CI 0.11-0.42), and concordance between aKi-67 and ORS was poor (κ 0.10, 95% CI -0.03-0.23).

CONCLUSIONS

These results highlight the limits of Ki-67 algorithms that use manual "hot spot" selection. Due to suboptimal concordance, Ki-67 is likely most useful as a complement to, rather than a surrogate for ORS, regardless of scoring method.

Solitary fibrous tumor of thoracic cavity, extra-thoracic sites and central nervous system: Clinicopathologic features and association with local recurrence and metastasis

Published risk stratification models of solitary fibrous tumor (SFT) have been associated with distant metastases outside the central nervous system (CNS), but have not been studied for tumors occurring in the CNS. In a retrospective review, we identified 72 cases of solitary fibrous tumor or hemangiopericytoma (HPC) diagnosed between January 2011 and December 2020 at our institution. The tumors involved the central nervous system (N = 17), thoracic cavity (N = 28), and extrathoracic sites (N = 27). The risk of local recurrence, distant metastasis, or death at 5 years was 57% (95% CI 23%, 76%) in the CNS, 24% (95% CI 2%, 41%) in the thoracic cavity, and 13% (95% CI 0%, 25%) in extrathoracic sites. By contrast, the risk of distant metastasis or death at 5 years was 13% (95% CI 0%, 29%) in CNS primaries, 5% (95% CI 0%, 14%) in thoracic primaries, and 14% (95% CI 0%, 27%) in extrathoracic primaries. Using the published 3- and 4-variable risk stratification models by Demicco et al., we retrospectively assessed our cases for risk of local recurrence, distant metastasis, and death. For tumors outside the CNS, we show that three- and four-variable risk stratification models were associated with recurrence-free survival in addition to the previously known association with distant metastasis (all P < 0.05). In contrast, inside the CNS, we show that neither risk model is a significantly associated with clinical behavior, and that WHO grade is likely the best available prognostic tool, though none of the differences were significant. The lack of significant differences can be likely explained by the younger median age (47 years vs 61 years) and smaller median tumor size (3.5 cm vs 5.6 cm), downgrading the risk stratification scores in CNS compared to non-CNS primaries. In conclusion, existing risk stratification models of SFT are not associated with clinical behavior for tumors arising inside the CNS, but are associated with local recurrence in addition to distant metastasis outside the CNS.

Integrated histologic and molecular analysis of uterine leiomyosarcoma and 2 benign variants with nuclear atypia

Uterine leiomyosarcoma (LMS) is a rare but deadly disease. Due to poor understanding of the molecular and genetic causes of the disease, the diagnosis of LMS has been based primarily on histology. Nuclear atypia is one of hallmarks in LMS, however, it also occurs in 2 clinically benign variants, including smooth muscle tumors with fumarate hydratase alteration (SMT‐FH) and leiomyoma with bizarre nuclei (LM‐BN). In addition to nuclear atypia, many well recognized biomarkers used for LMS are also frequently overexpressed in LM‐BN, and the histogenesis and molecular natures for LM‐BN and LMS remain largely unknown. To characterize the molecular profiling of LMS, SMT‐FH, and LM‐BN, we performed integrated comprehensive genomic profiling including whole‐genome sequencing (WGS) and RNA sequencing and genomic microarray analyses to assess genome‐wide copy number alterations (CNAs) and immunohistochemistry (IHC) in all 3 tumor types. We found that both LM‐BN and LMS showed genomic instability and harbored extensive CNAs throughout the whole genome. By contrast, the SMT‐FH presented its characteristic 1q43‐44 deletions in all cases tested, with minimal CNAs in the rest of genomic regions. Further analyses revealed that LMS and LM‐BN groups showed similar patterns of CNAs that are tended to cluster together and separated from the SMT‐FH group. The integrated molecular profiling enabled the detection of novel and traditional biomarkers and showed excellent discrimination between LM‐BN and LMS. Our study suggests that LM‐BN, despite having similar nuclear atypia to SMT‐FH, showed similar genomic instability but distinct genomic alterations with its malignant counterpart of LMS. The integrated molecular profiling is of clinical importance in characterizing these rare uterine smooth muscle tumors.

Oropharyngeal carcinoma: A single institution study of 338 primaries with special reference to high-risk human papillomavirus-mediated carcinoma with aggressive behavior

In a retrospective review, we identified 332 patients with 338 pathologically diagnosed primary oropharyngeal carcinomas (OPC) between January 2013 and March 2020 with known p16/HPV status from a tumor registry at Northwestern Memorial Hospital. The tumors predominantly involved the palatine tonsil (51 %) and the base of the tongue/lingual tonsil (38 %). The most common type of cancer was non-keratinizing squamous cell carcinoma (60 %), and the majority of primaries were p16 positive/HPV-mediated (86 %). A cohort of p16 positive/HPV mediated OPC (27/283, 9.5 %) presented with aggressive clinical behavior, including multiple distant metastases at unusual sites. Tumor size >2 cm and the presence of tumor anaplasia/multinucleation were significantly associated with an increased rate of distant metastases in p16 positive/HPV mediated cases, both in unadjusted and adjusted analyses (all P < 0.05). Of the 332 individuals in the overall cohort, 38 individuals died due to their disease within the observed follow-up time. Among the 283 patients with p16 positive/HPV mediated tumors, survival was estimated at 97 % (95 % CI 95 %, 100 %) at 1 year, 95 % (95 % CI 92 %, 98 %) at 2 years, and 80 % (95 % CI 72 %, 89 %) at 5 years. The presence of tumor anaplasia/multinucleation and distant metastasis were both significantly associated with poorer disease-specific survival in p16 positive/HPV mediated cases (both P < 0.05), with the survival effect of tumor anaplasia/multinucleation likely mediated in part through its association with distant metastasis. For p16 positive/HPV-mediated OPC, age, smoking status, tumor status, and lymph node status were not significantly associated with disease-specific survival in our study.

The diagnostic utility of PRAME and p16 in distinguishing nodal nevi from nodal metastatic melanoma

The status of the sentinel lymph node is the strongest predictor of recurrence in patients with malignant melanoma, making accurate distinction between nodal metastases and nodal nevi of paramount importance. We explored the utility of p16 and PRAME in differentiating nodal nevi from metastatic melanoma by immunohistochemistry. We searched our institutional database for cases of nodal nevi and nodal metastatic melanoma. p16 and PRAME expression were assessed with immunolabeling quantified by extent of nuclear positivity (0-25 %, >25 %-50 %, >50 %-75 % and >75 %). Sensitivities and specificities were calculated, and discrimination assessed using the area under the receiver operating characteristic curve (AUC). Forty-nine cases out of 51 nevi and 56/56 melanoma cases had lesional tissue present for p16, while 44/51 nevi and 54/56 melanoma cases had lesional tissue present for PRAME. 43 nodal nevi (88 %) had >50 % nuclear staining for p16, while none had >50 % staining for PRAME. More than half (55 %) of melanoma cases had complete loss of nuclear staining for p16, while majority (94 %) had >50 % nuclear staining for PRAME. Using a cut-off value of 50 %, higher PRAME expression had a sensitivity and specificity of 94 % and 100 %, respectively, while lower p16 expression had a sensitivity and specificity of 66 % and 88 %, respectively, for detecting metastatic melanoma. PRAME showed significantly better discrimination (AUC = 0.97, 95 % CI 0.94-1.00) than p16 (AUC = 0.77, 95 % CI 0.68-0.86) for differentiating nodal nevi from nodal melanoma (P < 0.001). Our findings suggest that PRAME is more accurate than p16 in discriminating between the two entities, with excellent sensitivity and specificity.

Detailed phenotyping reveals distinct trajectories of cardiovascular function and symptoms with exposure to modern breast cancer therapy

BACKGROUND

Breast cancer therapies are associated with a risk of cardiac dysfunction, most commonly defined by changes in left ventricular ejection fraction (LVEF). Recently, the authors identified 3 classes of LVEF change after exposure to anthracyclines and/or trastuzumab using latent class growth modeling. The objective of the current study was to characterize the clinical, biochemical, and functional profiles associated with LVEF trajectory class membership.

METHODS

Transthoracic echocardiography and biomarker assessments were performed and questionnaires were administered at standardized intervals in a longitudinal cohort of 314 patients with breast cancer who were treated with anthracyclines and/or trastuzumab. Univariable and multivariable multinomial regression analyses evaluated associations between baseline variables and LVEF trajectory class membership. Generalized estimating equations were used to define mean changes in cardiovascular measures over time within each class.

RESULTS

Among the 3 distinct subgroups of LVEF changes identified (stable [class 1]; modest, persistent decline [class 2]; and significant early decline followed by partial recovery [class 3]), higher baseline LVEF, radiotherapy, and sequential therapy with anthracyclines and/or trastuzumab were associated with class 2 or 3 membership. Sustained abnormalities in longitudinal strain and N-terminal pro-B-type natriuretic peptide (NT-proBNP) were observed in patients in class 2, as were heart failure symptoms. Similar abnormalities were observed in patients in class 3, but there was a trend toward recovery, particularly for longitudinal strain.

CONCLUSIONS

Patients with modest, persistent LVEF declines experienced sustained abnormalities in imaging and biochemical markers of cardiac function and heart failure symptoms. Further investigation is needed to characterize the long-term risk of heart failure, particularly in those with modest LVEF declines.

Abstract P4-02-09: Clinical and histopathologic characteristics of breast cancer in very young patients

Background: While breast cancer risk increases with age, about 7% of breast cancer cases in the US are diagnosed in women &lt;40 years of age. Evidence suggests that young breast cancer patients tend to have more aggressive disease subtypes, less favorable tumor biomarker profiles, higher risk of relapse, and poorer survival. However, there is no universally accepted definition of "young patients," and most studies have focused on premenopausal women or those &lt;40 years of age, with some studies using age &lt;35. To date, very little is known on the clinical and histopathologic characteristics of breast cancer in very young women, namely those ≤30 years of age. In this study, we describe such features in a cohort of very young women treated for breast cancer at our academic institution.

Methods: Our patient population consisted of all cases of invasive breast carcinoma at Northwestern Memorial Hospital in women &lt;40 years of age at surgery between January 1, 2009, and December 31, 2015. Very young was defined as having an age ≤30. Histopathologic features including tumor size, grade, histologic type, presence of lymphovascular invasion (LVI), lymph node status, tumor markers (ER, PR, HER2) and Ki-67 proliferation rate were recorded. Cases in which specific histopathologic features could not be determined from available materials were excluded from the corresponding analysis. The use of neoadjuvant chemotherapy, type of surgery, and the use of prophylactic surgery on the contralateral side were also recorded and analyzed.

Results: A total of 301 invasive breast carcinoma cases were identified (age range 18-39). 40 cases of very young (age ≤30) women with breast cancer (13%) were identified. Nearly all very young patients were diagnosed with tumors of ductal histology (39/40, 98%). Just under half of these patients (19/40, 48%) had a T1 (&lt;2 cm), 16/40 (40%) had a T2 (2-5 cm), and 2/40 (5%) had a T3 (&gt;5 cm) tumor at the time of surgery. The majority of these carcinomas were grade 3 (25/39, 64%), 14/39 (36%) were grade 2, and none were grade 1 tumors. Over half of the tumors were highly proliferative with a high Ki-67 count of &gt;20% (17/31, 55%). ER was expressed in 34/40 (85%), PR in 25/40 (63%) and HER2 in 9/40 (23%). LVI was present in half of the cases (20/40, 50%), and positive lymph nodes were identified in over a third (15/40, 38%) of the cases. Neoadjuvant chemotherapy was used in 9/40 (23%). Three patients (8%) had no residual carcinoma at the time of surgery post chemotherapy. Most of these very young patients chose a mastectomy (28/40, 70%), and 19/40 (48%) also underwent a prophylactic mastectomy on the contralateral side.

Conclusions: Our study is one of the largest to date to describe the characteristics of breast carcinomas in very young women (age ≤30). Poor pathologic features such as high tumor grade, high proliferation rate, and presence of LVI were seen in half or more of these very young patients. A high rate of positive lymph nodes was also seen. Mastectomy, often with contralateral prophylactic surgery, was the procedure of choice. Further studies evaluating the molecular characteristics of these tumors and the prevalence of underlying genetic mutations are under way in this unique population of very young breast cancer patients.

Citation Format: Finkelman BS, Blanco LZ, Siziopikou KP. Clinical and histopathologic characteristics of breast cancer in very young patients [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P4-02-09.

Abnormalities in 3-Dimensional Left Ventricular Mechanics With Anthracycline Chemotherapy Are Associated With Systolic and Diastolic Dysfunction

OBJECTIVES

The objective of this study was to evaluate the changes in three-dimensional (3D) speckle-tracking echocardiography-derived measures of mechanics and their associations with systolic and diastolic dysfunction after anthracyclines.

BACKGROUND

An improved understanding of the changes in 3D cardiac mechanics with anthracyclines may provide important mechanistic insight and identify new metrics to detect cardiac dysfunction.

METHODS

A total of 142 women with breast cancer receiving doxorubicin (240 mg/m²) with or without trastuzumab underwent 3D speckle-tracking echocardiography at standardized intervals prior to, during, and annually after chemotherapy. Left ventricular ejection fraction (LVEF), global circumferential strain (GCS), global longitudinal strain (GLS), principal strain, twist, and torsion were quantified. Linear regression analyses defined the associations between clinical factors and 3D parameters. Linear regression models with cluster robust variance estimators determined the associations between 3D measures and 2-dimensional (2D) LVEF and Doppler-derived E/e' over time.

RESULTS

There were significant abnormalities in 3D LVEF, GCS, GLS, and principal strain post-doxorubicin compared with control subjects (p < 0.001). The 3D parameters worsened post-anthracyclines, and only partially recovered to baseline over a median of 2.1 years (interquartile range: 1 to 4 years). Higher blood pressure and body mass index were associated with worse post-anthracycline 3D GCS and GLS, respectively. All 3D measures were associated with 2D LVEF at the same visit; only 3D LVEF, GCS, GLS, and principal strain were associated with 2D LVEF at subsequent visits (p < 0.05). In exploratory analyses, 3D LVEF and GCS were associated with subsequent systolic function independent of their corresponding 2D measures. The 3D LVEF, GCS, principal strain, and twist were significantly associated with concurrent, but not subsequent, E/e'.

CONCLUSIONS

Anthracyclines result in early and persistent abnormalities in 3D mechanics. The 3D LVEF and strain measures are associated with concurrent and subsequent systolic dysfunction, and concurrent diastolic dysfunction. Future research is needed to define the mechanisms and clinical relevance of abnormal 3D mechanics.

Arginine-Nitric Oxide Metabolites and Cardiac Dysfunction in Patients With Breast Cancer

BACKGROUND

Oxidative/nitrosative stress and endothelial dysfunction are hypothesized to be central to cancer therapeutics-related cardiac dysfunction (CTRCD). However, the relationship between circulating arginine-nitric oxide (NO) metabolites and CTRCD remains unstudied.

OBJECTIVES

This study sought to examine the relationship between arginine-NO metabolites and CTRCD in a prospective cohort of 170 breast cancer patients treated with doxorubicin with or without trastuzumab.

METHODS

Plasma levels of arginine, citrulline, ornithine, asymmetric dimethylarginine (ADMA), symmetric dimethylarginine (SDMA), and N-monomethylarginine (MMA) were quantified at baseline, 1 month, and 2 months after doxorubicin initiation. Determinants of baseline biomarker levels were identified using multivariable linear regression, and Cox regression defined the association between baseline levels and 1- or 2-month biomarker changes and CTRCD rate in 139 participants with quantitated echocardiograms at all time points.

RESULTS

Age, hypertension, body mass index, and African-American race were independently associated with ≥1 of baseline citrulline, ADMA, SDMA, and MMA levels. Decreases in arginine and citrulline and increases in ADMA were observed at 1 and 2 months (all p < 0.05). Overall, 32 participants experienced CTRCD over a maximum follow-up of 5.4 years. Hazard ratios for ADMA and MMA at 2 months were 3.33 (95% confidence interval [CI]: 1.12 to 9.96) and 2.70 (95% CI: 1.35 to 5.41), respectively, and 0.78 (95% CI: 0.64 to 0.97) for arginine at 1 month.

CONCLUSIONS

In breast cancer patients undergoing doxorubicin therapy, early alterations in arginine-NO metabolite levels occurred, and early biomarker changes were associated with a greater CTRCD rate. Our findings highlight the potential mechanistic and translational relevance of this pathway to CTRCD.

A clinical prediction model to assess risk for pancreatic cancer among patients with new-onset diabetes

e15780

Background: Approximately 50% of all patients with pancreatic ductal adenocarcinoma (PDA) develop diabetes mellitus (DM) prior to cancer diagnosis. Targeted screening for PDA among those with new-onset diabetes may allow earlier diagnosis. We sought to develop and validate a PDA risk prediction model to identify high-risk individuals among those with new-onset diabetes. Methods: We conducted a retrospective cohort study in a population representative database from the UK. Individuals with incident diabetes after the age of 35 and ≥3 years of follow-up after DM diagnosis were eligible for inclusion. Candidate predictors consisted of epidemiological and clinical characteristics available at the time of diabetes diagnosis. Variables with p-value<0.25 in the univariable analyses were further evaluated using backward stepwise approach. Model discrimination was assessed using ROC curve analysis. Calibration was evaluated using the Hosmer–Lemeshow test. Results were internally validated using a bootstrapping procedure. Results: The study included 109,385 patients with new-onset diabetes. Among them, 390 (0.4%) were diagnosed with PDA within 3 years. The final model (AUC 0.82, 95% CI: 0.75-0.89) included age, BMI, BMI change, smoking, HbA1C, cholesterol, hemoglobin, creatinine and alkaline phosphatase, and use of PPI and anti-diabetic medication. Bootstrapping validation showed negligible optimism. If the predicted risk threshold for definitive PDA screening was set at 1% over 3 years, only 6.19% of the new-onset diabetes population would undergo definitive screening, and the corresponding sensitivity, specificity and positive predictive value would be 44.7%, 94.0%, and 2.6% respectively. Conclusions: A risk model based on widely available clinical parameters can help target PDA screening in patients with new-onset diabetes.

Prospective Evaluation of Sunitinib-Induced Cardiotoxicity in Patients with Metastatic Renal Cell Carcinoma

Purpose: To prospectively evaluate cardiotoxicity risk with sunitinib in metastatic renal cell carcinoma (mRCC) routine clinical practice using comprehensive echocardiography and biomarker phenotyping.Experimental Design: In a multicenter prospective study of 90 patients with mRCC, echocardiography and biomarkers of cardiovascular injury and stress were quantified at baseline, 3.5, 15, and 33 weeks following sunitinib initiation. These "on-drug" visits corresponded to cycles 1, 3, and 6, respectively. Left ventricular (LV) dysfunction was defined as an absolute decline in LV ejection fraction (LVEF) by ≥10% to a value of <50%. Conditional survival analyses predicted the risk of LV dysfunction. Linear mixed-effects models estimated changes in LVEF, high-sensitivity Troponin I (hsTnI), and B-type natriuretic peptide (BNP) over time.Results: The predicted risk of LV dysfunction by cycle 6 was 9.7% (95% confidence interval, 3%-17%). The majority of events occurred in the first treatment cycle. This risk diminished to 5% and 2% in patients who had not experienced dysfunction by the completion of cycles 1 and 3, respectively. All evaluable patients who experienced LV dysfunction had subsequent improvement in LVEF with careful management. Six patients (6.7%) developed hsTnI elevations >21.5 pg/mL, and 11 additional patients (12.2%) developed BNP elevations >100 pg/mL. These elevations similarly tended to occur early and resolved over time.Conclusions: On average, patients with mRCC receiving sunitinib exhibit modest declines in LVEF and nonsignificant changes in hsTnI and BNP. However, approximately 9.7% to 18.9% of patients develop more substantive abnormalities. These changes occur early and are largely recoverable with careful management. Clin Cancer Res; 23(14); 3601-9. ©2017 AACR.

Predicting prolonged dose titration in patients starting warfarin

PURPOSE

Patients initiating warfarin therapy generally experience a dose-titration period of weeks to months, during which time they are at higher risk of both thromboembolic and bleeding events. Accurate prediction of prolonged dose titration could help clinicians determine which patients might be better treated by alternative anticoagulants that, while more costly, do not require dose titration.

METHODS

A prediction model was derived in a prospective cohort of patients starting warfarin (n = 390), using Cox regression, and validated in an external cohort (n = 663) from a later time period. Prolonged dose titration was defined as a dose-titration period >12 weeks. Predictor variables were selected using a modified best subsets algorithm, using leave-one-out cross-validation to reduce overfitting.

RESULTS

The final model had five variables: warfarin indication, insurance status, number of doctor's visits in the previous year, smoking status, and heart failure. The area under the ROC curve (AUC) in the derivation cohort was 0.66 (95%CI 0.60, 0.74) using leave-one-out cross-validation, but only 0.59 (95%CI 0.54, 0.64) in the external validation cohort, and varied across clinics. Including genetic factors in the model did not improve the area under the ROC curve (0.59; 95%CI 0.54, 0.65). Relative utility curves indicated that the model was unlikely to provide a clinically meaningful benefit compared with no prediction.

CONCLUSIONS

Our results suggest that prolonged dose titration cannot be accurately predicted in warfarin patients using traditional clinical, social, and genetic predictors, and that accurate prediction will need to accommodate heterogeneities across clinical sites and over time. Copyright © 2016 John Wiley & Sons, Ltd.

Idiopathic multicentric Castleman's disease: a systematic literature review

BACKGROUND

Multicentric Castleman's disease describes a group of poorly understood lymphoproliferative disorders driven by proinflammatory hypercytokinaemia. Patients have heterogeneous clinical features, characteristic lymph node histopathology, and often deadly multiple organ dysfunction. Human herpesvirus 8 (HHV8) causes multicentric Castleman's disease in immunosuppressed patients. The cause of HHV8-negative multicentric Castleman's disease is idiopathic; such cases are called idiopathic multicentric Castleman's disease. An absence of centralised information about idiopathic multicentric Castleman's disease represents a major challenge for clinicians and researchers. We aimed to characterise clinical features of, treatments for, and outcomes of idiopathic multicentric Castleman's disease.

METHODS

We did a systematic literature review and searched PubMed, the Cochrane database, and ClinicalTrials.gov from January, 1995, with keywords including "Castleman's disease" and "giant lymph node hyperplasia". Inclusion criteria were pathology-confirmed Castleman's disease in multiple nodes and minimum clinical and treatment information on individual patients. Patients with HHV8 or HIV infection or diseases known to cause Castleman-like histopathology were excluded.

FINDINGS

Our search identified 626 (33%) patients with HHV8-negative multicentric Castleman's disease from 1923 cases of multicentric Castleman's disease. 128 patients with idiopathic multicentric Castleman's disease met all inclusion criteria for the systematic review. Furthermore, aggregated data for 127 patients with idiopathic multicentric Castleman's disease were presented from clinical trials, which were excluded from primary analyses because patient-level data were not available. Clinical features of idiopathic multicentric Castleman's disease included multicentric lymphadenopathy (128/128), anaemia (79/91), elevated C-reactive protein (65/79), hypergammaglobulinaemia (63/82), hypoalbuminaemia (57/63), elevated interleukin 6 (57/63), hepatomegaly or splenomegaly (52/67), fever (33/64), oedema, ascites, anasarca, or a combination (29/37), elevated soluble interleukin 2 receptor (20/21), and elevated VEGF (16/20). First-line treatments for idiopathic multicentric Castleman's disease included corticosteroids (47/128 [37%]), cytotoxic chemotherapy (47/128 [37%]), and anti-interleukin 6 therapy (11/128 [9%]). 49 (42%) of 116 patients failed first-line therapy, 2-year survival was 88% (95% CI 81-95; 114 total patients, 12 events, 36 censored), and 27 (22%) of 121 patients died by the end of their observed follow-up (median 29 months [IQR 12-50]). 24 (19%) of 128 patients with idiopathic multicentric Castleman's disease had a diagnosis of a separate malignant disease, significantly higher than the frequency expected in age-matched controls (6%).

INTERPRETATION

Our systematic review provides comprehensive information about clinical features, treatment, and outcomes of idiopathic multicentric Castleman's disease, which accounts for at least 33% of all cases of multicentric Castleman's disease. Our findings will assist with prompt recognition, diagnostic criteria development, and effective management of the disease.

FUNDING

None.

The prediction accuracy of dynamic mixed-effects models in clustered data

BACKGROUND

Clinical prediction models often fail to generalize in the context of clustered data, because most models fail to account for heterogeneity in outcome values and covariate effects across clusters. Furthermore, standard approaches for modeling clustered data, including generalized linear mixed-effects models, would not be expected to provide accurate predictions in novel clusters, because such predictions are typically based on the hypothetical mean cluster. We hypothesized that dynamic mixed-effects models, which incorporate data from previous predictions to refine the model for future predictions, would allow for cluster-specific predictions in novel clusters as the model is updated over time, thus improving overall model generalizability.

RESULTS

We quantified the potential gains in prediction accuracy from using a dynamic modeling strategy in a simulation study. Furthermore, because clinical prediction models in the context of clustered data often involve outcomes that are dependent on patient volume, we examined whether using dynamic mixed-effects models would be robust to misspecification of the volume-outcome relationship. Our results indicated that dynamic mixed-effects models led to substantial improvements in prediction accuracy in clustered populations over a broad range of conditions, and were uniformly superior to static models. In addition, dynamic mixed-effects models were particularly robust to misspecification of the volume-outcome relationship and to variation in the frequency of model updating. The extent of the improvement in prediction accuracy that was observed with dynamic mixed-effects models depended on the relative impact of fixed and random effects on the outcome as well as the degree of misspecification of model fixed effects.

CONCLUSIONS

Dynamic mixed-effects models led to substantial improvements in prediction model accuracy across a broad range of simulated conditions. Therefore, dynamic mixed-effects models could be a useful alternative to standard static models for improving the generalizability of clinical prediction models in the setting of clustered data, and, thus, well worth the logistical challenges that may accompany their implementation in practice.

Factors affecting time to maintenance dose in patients initiating warfarin

PURPOSE

Patients starting warfarin often experience lengthy dose-titration periods, when they are at high risk for bleeding and thromboembolism. However, relatively little is known about why some patients take longer than others to reach maintenance dose. Thus, we sought to identify social, clinical, and genetic factors associated with prolonged time to maintenance dose (TTM).

METHODS

We conducted a time-to-event analysis, using a prospective cohort of patients initiating warfarin (N = 390). Additionally, we examined whether changes in post-initiation factors were associated with TTM. Finally, we performed a secondary analysis in a subcohort (N = 156) assessing the effect of adherence on TTM.

RESULTS

No genetic or post-initiation factors were significantly associated with TTM. However, previous use of warfarin [hazard ratio (HR) = 0.64; 95% confidence interval (CI) 0.46, 0.88], current smoking status (HR = 0.61; 95%CI 0.39, 0.96), fewer than four doctor's visits in the previous year (HR = 0.63 vs 4-12 visits; 95%CI 0.46, 0.88), and worse general health status (HR = 0.63; 95%CI 0.47, 0.84) were significantly associated with longer TTM. Use of illegal injectable drugs (HR = 2.51; 95%CI 1.17, 5.39) was associated with shorter TTM. On secondary analysis, the HR for better adherence and TTM was 1.70 (95%CI 0.88, 3.27).

CONCLUSIONS

Time to maintenance dose was associated with pre-existing behavioral factors, health care utilization, and health quality but not clinical comorbidities or genetic factors in patients initiating warfarin. Future studies are needed to determine whether warfarin patients with prolonged TTM would have better outcomes on alternative agents.

Incidence of bladder cancer in patients with type 2 diabetes treated with metformin or sulfonylureas

OBJECTIVE

Previous studies evaluating the effect of metformin on cancer risk have been impacted by time-related biases. To avoid these biases, we examined the incidence of bladder cancer in new users of metformin and sulfonylureas (SUs).

RESEARCH DESIGN AND METHODS

This cohort study included 87,600 patients with type 2 diabetes in The Health Improvement Network database. Use of metformin or an SU was treated as a time-dependent variable. Cox regression-generated hazard ratios (HRs) compared metformin use with SU use, adjusted for age, sex, smoking, obesity, and HbA1c level.

RESULTS

We identified 196 incident bladder cancers in the metformin cohort and 66 cancers in the SU cohort. Use of metformin was not associated with decreased bladder cancer risk (HR 0.81 [95% CI 0.60-1.09]). This association did not differ by sex (P for interaction = 0.20). We observed no association with duration of metformin relative to SU use (3 to <4 years of use: 0.57 [0.25-1.34]; 4 to <5 years of use: 0.93 [0.30-2.85; ≥5 years of use: 1.18 [0.44-3.19]; P for trend = 0.26).

CONCLUSIONS

Use of metformin is not associated with a decreased incidence of bladder cancer. Similar methods should be used to study other cancers that have previously been identified as potentially preventable with metformin.

Distinguishing incident and prevalent diabetes in an electronic medical records database

PURPOSE

To develop a method to identify incident diabetes mellitus (DM) using an electronic medical records (EMR) database and test this classification by comparing incident and prevalent DM with common outcomes related to DM duration.

METHODS

Incidence rates (IRs) of DM (defined as a first diagnosis or prescription) were measured in 3-month intervals through 36 months after registration in The Health Improvement Network, a primary care database, from 1994 to 2012. We used Joinpoint regression to identify the point where a statistically significant change in the trend of IRs occurred. Further analyses used this point to distinguish those likely to have incident (n = 50 315) versus prevalent (n = 28 337) DM. Incident and prevalent cohorts were compared using Cox regression for all-cause mortality, cardiovascular disease (CVD), diabetic retinopathy, diabetic nephropathy, and diabetic neuropathy. Analyses were adjusted for age, sex, smoking, obesity, hyperlipidemia, hypertension, and calendar year.

RESULTS

Trends in DM IRs plateaued 9 months after registration (p = 0.04). All cause-mortality was increased (hazard ratio (HR) 1.62, 95% CI 1.53-1.70) among patients diagnosed with DM prior to 9 months following registration (prevalent DM) compared to those diagnosed after 9 months (incident DM). Similarly, the risk of DM-related complications was higher in prevalent versus incident DM patients [CVD, HR 2.24 (2.08-2.40); diabetic retinopathy, HR 1.31 (1.24-1.38); diabetic nephropathy, HR 2.30 (1.95-2.72); diabetic neuropathy, HR 1.28 (1.16-1.41)].

CONCLUSION

Joinpoint regression can be used to identify patients with newly diagnosed diabetes within EMR data. Failure to exclude patients with prevalent DM can lead to exaggerated associations of DM-related outcomes.

Disparities between high-impact clinical trials in oncology and disease burden

e17591

Background: Recent research has identified a discrepancy between NIH funding in oncology and various measures of disease burden by tumor type. We sought to identify whether this disparity exists in recent high impact publications. Methods: 833 clinical trials published in five high impact general medicine and clinical oncology journals between January, 2009 and October, 2012 were reviewed. 692 trials were included in this analysis after excluding those that studied >1 tumor type. Disease burden was measured as person-years of life lost (YLLs), reported in the Surveillance, Epidemiology, and End Results database, and disability adjusted life years (DALYs), reported by the World Health Organization. We used a chi square goodness of fit test to compare the overall distribution of trials by tumor type to the distribution of annual YLLs and DALYs. Results: Breast cancer was the most published tumor, accounting for 14% of all trials, followed by lung (13%) and colorectal (7%) cancers. More than half of the trials (56%) were for patients with metastatic disease, and most (81%) were phase 2 and 3 clinical trials. Nearly half of all publications studied targeted therapies (45%), and the majority received industry support (61%). 67% of trials with a comparator arm met their primary endpoint. The distribution of trials by cancer site differed significantly from the distribution of both measures of disease burden (YLLs and DALYs) (both p<0.001). The findings were unchanged in analyses that accounted for the total number of subjects enrolled in the trials (both p<0.001). The most underrepresented malignancies based on burden of disease were lung and pancreatic cancers, while the most overrepresented were breast cancer, leukemia, and melanoma. Conclusions: The number of trials published by tumor type does not directly reflect the burden of these diseases in the population as assessed by YLL or DALY. Future studies examining potential confounders such as funding availability by cancer type or number of unpublished clinical trials may further clarify the observed disparities.

Breast and ovarian cancer risk and risk reduction in Jewish BRCA1/2 mutation carriers

PURPOSE

Mutations in BRCA1/2 dramatically increase the risk of both breast and ovarian cancers. Three mutations in these genes (185delAG, 5382insC, and 6174delT) occur at high frequency in Ashkenazi Jews. We evaluated how these common Jewish mutations (CJMs) affect cancer risks and risk reduction.

METHODS

Our cohort comprised 4,649 women with disease-associated BRCA1/2 mutations from 22 centers in the Prevention and Observation of Surgical End Points Consortium. Of these women, 969 were self-identified Jewish women. Cox proportional hazards models were used to estimate breast and ovarian cancer risks, as well as risk reduction from risk-reducing salpingo-oophorectomy (RRSO), by CJM and self-identified Jewish status.

RESULTS

Ninety-one percent of Jewish BRCA1/2-positive women carried a CJM. Jewish women were significantly more likely to undergo RRSO than non-Jewish women (54% v 41%, respectively; odds ratio, 1.87; 95% CI, 1.44 to 2.42). Relative risks of cancer varied by CJM, with the relative risk of breast cancer being significantly lower in 6174delT mutation carriers than in non-CJM BRCA2 carriers (hazard ratio, 0.35; 95% CI, 0.18 to 0.69). No significant difference was seen in cancer risk reduction after RRSO among subgroups.

CONCLUSION

Consistent with previous results, risks for breast and ovarian cancer varied by CJM in BRCA1/2 carriers. In particular, 6174delT carriers had a lower risk of breast cancer. This finding requires additional confirmation in larger prospective and population-based cohort studies before being integrated into clinical care.

Genetic warfarin dosing: tables versus algorithms

OBJECTIVES

The aim of this study was to compare the accuracy of genetic tables and formal pharmacogenetic algorithms for warfarin dosing.

BACKGROUND

Pharmacogenetic algorithms based on regression equations can predict warfarin dose, but they require detailed mathematical calculations. A simpler alternative, recently added to the warfarin label by the U.S. Food and Drug Administration, is to use genotype-stratified tables to estimate warfarin dose. This table may potentially increase the use of pharmacogenetic warfarin dosing in clinical practice; however, its accuracy has not been quantified.

METHODS

A retrospective cohort study of 1,378 patients from 3 anticoagulation centers was conducted. Inclusion criteria were stable therapeutic warfarin dose and complete genetic and clinical data. Five dose prediction methods were compared: 2 methods using only clinical information (empiric 5 mg/day dosing and a formal clinical algorithm), 2 genetic tables (the new warfarin label table and a table based on mean dose stratified by genotype), and 1 formal pharmacogenetic algorithm, using both clinical and genetic information. For each method, the proportion of patients whose predicted doses were within 20% of their actual therapeutic doses was determined. Dosing methods were compared using McNemar's chi-square test.

RESULTS

Warfarin dose prediction was significantly more accurate (all p < 0.001) with the pharmacogenetic algorithm (52%) than with all other methods: empiric dosing (37%; odds ratio [OR]: 2.2), clinical algorithm (39%; OR: 2.2), warfarin label (43%; OR: 1.8), and genotype mean dose table (44%; OR: 1.9).

CONCLUSIONS

Although genetic tables predicted warfarin dose better than empiric dosing, formal pharmacogenetic algorithms were the most accurate.

New genetic variant that might improve warfarin dose prediction in African Americans

WHAT IS ALREADY KNOWN ABOUT THIS SUBJECT

Variants in the CYP2C9 (i.e. *2 and *3) and VKORC1 (i.e. 1173C/T or -1639G/A) genes have been shown to influence warfarin dose requirements. However, these factors seem to explain less of the dose variability in African Americans who have a lower prevalence of the CYP2C9*2 and *3 and VKORC1 1173T alleles.

WHAT THIS STUDY ADDS

In African Americans, the VKORC1 rs17886199 variant was statistically significantly associated with log-transformed warfarin maintenance dose, independent of the influence of VKORC1 1173C>T and CYP2C9*2 and *3. However, replication of our finding is needed to confirm the association of rs1786199 SNP in African Americans, since Limdi et al.[3] did not examine the effect of this SNP because the prevalence of the rs1786199 A-allele was too low.

AIMS

To raise hypotheses with regards to whether genetic variants in the VKORC1, CYP2C9, EPHX1, GGCX and ALB genes might influence warfarin dose in African Americans and Caucasians, independent of the effects of the VKORC1 1173C>T and CYP2C9*2 and *3 variants.

METHODS

From a prospective cohort study, we obtained additional DNA on 36 Caucasian and 22 African American warfarin users who reached maintenance dose and genotyped them for tagSNPs (r2<0.8) in VKORC1, EPHX1, GGCX and ALB genes, and one exonic CYP2C9 SNP. Linear regression models were fitted to estimate the relationship (P value) between log-transformed maintenance dose and each SNP and the amount of the warfarin dose variability accounted for by each SNP (partial R2).

RESULTS

In African Americans, the VKORC1 rs17886199 A-allele was associated with a lower dose (GG=46.3 mg and GA=25.6 mg; P=0.002), independent of the VKORC1 1173C>T and CYP2C9*2 and *3 variants. Even after applying Bonferroni correction, the P value would still be considered statistically significant. The VKORC1 rs17886199 variant was not found in Caucasians. In Caucasians, the EPHX1 rs1051741 T-allele was associated with a lower dose (CC=41.3 mg and CT=30.0 mg; P=0.04). The latter was no longer statistically significant after applying Bonferroni correction.

CONCLUSIONS

Our pilot study suggests that the VKORC1 rs17886199 variant could influence warfarin maintenance dose among African Americans, even after accounting for the influence of the VKORC1 1173C>T variant. Future studies with a larger sample size will be needed to confirm our findings.

Clinical Factors Associated with Bacterial Biofilm Formation in Chronic Rhinosinusitis

Objectives. Bacterial biofilms appear to contribute to chronic rhinosinusitis. However, the mechanism behind biofilm formation in chronic rhinosinusitis remains poorly defined. The aim of this study is to evaluate clinical factors that may be associated with bacterial biofilm formation in chronic rhinosinusitis.

Study Design. Cross-sectional study.

Setting. Department of Otorhinolaryngology–Head and Neck Surgery at the Hospital of the University of Pennsylvania.

Subjects and Methods. Five hundred eighteen patients with chronic rhinosinusitis were enrolled from 2007 to 2010. Samples were taken to evaluate for biofilm formation in vitro using a modified Calgary Biofilm Detection Assay. Clinical data were collected from chart review. Pearson’s χ2 and logistic regression were used for the analyses.

Results. Of the patients, 108 (20.9%) showed biofilm formation in vitro. Bacterial biofilm formation in vitro was not significantly associated with polyps, allergy, Samter’s triad, sleep apnea, smoking status, age, or gender. However, it was significantly associated with positive culture results (odds ratio [OR] = 3.13; 95% confidence interval [CI], 1.85-5.29; P < .001), prior sinus surgeries (1.93; 1.01-3.69; P = .046), and nasal steroid use in the month prior to sample collection (2.09; 1.07-4.08; P = .030). Polymicrobial cultures, Pseudomonas aeruginosa, and Staphylococcus aureus comprised most of the samples.

Conclusion. The results of this study suggest that the probability of bacterial biofilm formation is independent of many clinical factors considered to be risk factors for chronic rhinosinusitis. Further studies are needed to clarify the nature of the associations between prior sinus surgeries, nasal steroid use, and biofilm formation.

In sickness and in health care: a student's thoughts before beginning his medical training

This essay is a "prespective"-the musings of a soon-to-be MD/PhD student on various aspects of the practice of medicine, written just before he began his training in the fall of 2008. It discusses some of the issues-genomic medicine, healthcare reform, and evidence-based medicine-that will likely impact medicine and medical care during his career. These thoughts are interwoven with the personal story of his grandfather's fight against disease and the complications of diagnosis and treatment.

Global patterns in seasonal activity of influenza A/H3N2, A/H1N1, and B from 1997 to 2005: viral coexistence and latitudinal gradients

Despite a mass of research on the epidemiology of seasonal influenza, overall patterns of infection have not been fully described on broad geographic scales and for specific types and subtypes of the influenza virus. Here we provide a descriptive analysis of laboratory-confirmed influenza surveillance data by type and subtype (A/H3N2, A/H1N1, and B) for 19 temperate countries in the Northern and Southern hemispheres from 1997 to 2005, compiled from a public database maintained by WHO (FluNet). Key findings include patterns of large scale co-occurrence of influenza type A and B, interhemispheric synchrony for subtype A/H3N2, and latitudinal gradients in epidemic timing for type A. These findings highlight the need for more countries to conduct year-round viral surveillance and report reliable incidence data at the type and subtype level, especially in the Tropics.