MoViT: Memorizing Vision Transformers for Medical Image Analysis

The synergy of long-range dependencies from transformers and local representations of image content from convolutional neural networks (CNNs) has led to advanced architectures and increased performance for various medical image analysis tasks due to their complementary benefits. However, compared with CNNs, transformers require considerably more training data, due to a larger number of parameters and an absence of inductive bias. The need for increasingly large datasets continues to be problematic, particularly in the context of medical imaging, where both annotation efforts and data protection result in limited data availability. In this work, inspired by the human decision-making process of correlating new "evidence" with previously memorized "experience", we propose a Memorizing Vision Transformer (MoViT) to alleviate the need for large-scale datasets to successfully train and deploy transformer-based architectures. MoViT leverages an external memory structure to cache history attention snapshots during the training stage. To prevent overfitting, we incorporate an innovative memory update scheme, attention temporal moving average, to update the stored external memories with the historical moving average. For inference speedup, we design a prototypical attention learning method to distill the external memory into smaller representative subsets. We evaluate our method on a public histology image dataset and an in-house MRI dataset, demonstrating that MoViT applied to varied medical image analysis tasks, can outperform vanilla transformer models across varied data regimes, especially in cases where only a small amount of annotated data is available. More importantly, MoViT can reach a competitive performance of ViT with only 3.0% of the training data. In conclusion, MoViT provides a simple plug-in for transformer architectures which may contribute to reducing the training data needed to achieve acceptable models for a broad range of medical image analysis tasks.

CEST2022: Amide proton transfer-weighted MRI improves the diagnostic performance of multiparametric non-contrast-enhanced MRI techniques in patients with post-treatment high-grade gliomas

New or enlarged lesions in malignant gliomas after surgery and chemoradiation can be associated with tumor recurrence or treatment effect. Due to similar radiographic characteristics, conventional-and even some advanced MRI techniques-are limited in distinguishing these two pathologies. Amide proton transfer-weighted (APTw) MRI, a protein-based molecular imaging technique that does not require the administration of any exogenous contrast agent, was recently introduced into the clinical setting. In this study, we evaluated and compared the diagnostic performances of APTw MRI with several non-contrast-enhanced MRI sequences, such as diffusion-weighted imaging, susceptibility-weighted imaging, and pseudo-continuous arterial spin labeling. Thirty-nine scans from 28 glioma patients were obtained on a 3 T MRI scanner. A histogram analysis approach was employed to extract parameters from each tumor area. Statistically significant parameters (P < 0.05) were selected to train multivariate logistic regression models to evaluate the performance of MRI sequences. Multiple histogram parameters, particularly from APTw and pseudo-continuous arterial spin labeling images, demonstrated significant differences between treatment effect and recurrent tumor. The regression model trained on the combination of all significant histogram parameters achieved the best result (area under the curve = 0.89). We found that APTw images added value to other advanced MR images for the differentiation of treatment effect and tumor recurrence.

Reconstruction of Large Anterior Skull Base Defects After Resection of Sinonasal Tumors With Intracranial Extension by Using Pedicled Double Flap Techniques

INTRODUCTION

The use of surgical resection for large anterior skull base (ASB) tumors and sinonasal malignancies with intracranial extension will result in a large skull base defect. Reconstruction of large ASB defects using traditional techniques is high risk and may lead to postoperative cerebral spinal fluid (CSF) leakage, meningitis, and an increase in mortality rate. The use of a pedicled double flap technique to reconstruct the ASB defect may decrease complications. This study presents the clinical outcomes of patients who underwent double flap reconstruction techniques after resection of their sinonasal malignancies with significant intracranial extension at Cho Ray hospital in Vietnam.

METHODS

The case series study was conducted at Cho Ray hospital from September 2010 to September 2020. All patients with large sinonasal malignancies that invaded intracranially underwent transnasal endoscopic surgery and subfrontal craniotomy. Reconstruction of large skull base defects (>2 cm) were followed up by using the pedicled double flaps technique. This study was performed in line with the principles of the Declaration of Helsinki. Approval of the study was granted by the Independent Ethics Committee of Cho Ray Hospital (Date: March 3, 2014/No: 11/BVCRHĐĐĐ).

RESULTS

During September 2010 to September 2020, there were 75 patients who underwent a modified multilayer, double flap reconstruction technique after the resection of their ASB tumor. Skull base defects were commonly seen along the horizontal plate of the ethmoid bone and the ethmoid roof (98.6%). Large skull base defects (>2 cm) accounted for 81.3% of cases. Overall, the risk of postoperative CSF leakage and meningitis after double flap repair was considerably low. Of all participants, only 1 experienced postoperative CSF leakage and 1 experienced postoperative meningitis. Despite the complications, these patients improved significantly and remained stable.

CONCLUSION

The use of double vascularized pedicled flaps may decrease the incidence of postoperative CSF leakage and meningitis. This technique is an effective method for the reconstruction of ASB tumors with large defects.

Generating Predictable and Adaptive Dialog Policies in Single- and Multi-domain Goal-oriented Dialog Systems

Most existing commercial goal-oriented chatbots are diagram-based; i.e. they follow a rigid dialog flow to fill the slot values needed to achieve a user’s goal. Diagram-based chatbots are predictable, thus their adoption in commercial settings; however, their lack of flexibility may cause many users to leave the conversation before achieving their goal. On the other hand, state-of-the-art research chatbots use Reinforcement Learning (RL) to generate flexible dialog policies. However, such chatbots can be unpredictable, may violate the intended business constraints, and require large training datasets to produce a mature policy. We propose a framework that achieves a middle ground between the diagram-based and RL-based chatbots: we constrain the space of possible chatbot responses using a novel structure, the chatbot dependency graph, and use RL to dynamically select the best valid responses. Dependency graphs are directed graphs that conveniently express a chatbot’s logic by defining the dependencies among slots: all valid dialog flows are encapsulated in one dependency graph. Our experiments in both single-domain and multi-domain settings show that our framework quickly adapts to user characteristics and achieves up to 23.77% improved success rate compared to a state-of-the-art RL model.

Global-local attention for emotion recognition

Human emotion recognition is an active research area in artificial intelligence and has made substantial progress over the past few years. Many recent works mainly focus on facial regions to infer human affection, while the surrounding context information is not effectively utilized. In this paper, we proposed a new deep network to effectively recognize human emotions using a novel global-local attention mechanism. Our network is designed to extract features from both facial and context regions independently, then learn them together using the attention module. In this way, both the facial and contextual information is used to infer human emotions, therefore enhancing the discrimination of the classifier. The intensive experiments show that our method surpasses the current state-of-the-art methods on recent emotion datasets by a fair margin. Qualitatively, our global-local attention module can extract more meaningful attention maps than previous methods. The source code and trained model of our network are available at https://github.com/minhnhatvt/glamor-net.

Trial leadership by women is associated with racial diversity among heart failure clinical trial participants: a systematic bibliometric review 2000–2020

Background

For the results of randomized controlled trials (RCTs) to be generalizable, they should report on and include the broad range of patients who have the disease.

Purpose

We assessed temporal trends and trial factors associated with 1) the reporting of race or ethnicity data and 2) the enrolment of Black, Indigenous, and people of colour (BIPOC) in Heart Failure (HF) RCTs.

Methods

We searched MEDLINE, EMBASE, and CINAHL for RCTs that recruited adults with HF and were published in journals with an impact factor ≥10 between January 1, 2000 and June 17, 2020. We extracted data in duplicate and used the Cochran-Armitage and Jonchkeere-Terpstra tests to examine temporal trends. We used multivariable regression to assess the independent association between trial factors and the outcomes of interest.

Results

A total of 414 RCTs met inclusion criteria, of which a vast majority (90.6%; 95% CI 87.4–93.2%) were coordinated in either Europe or North America. Only 157 of the 414 RCTs (37.9%; 95% CI 33.2–42.8%) reported race/ethnicity data; among the 158,200 participants in these trials, only 29,512 (18.7%; 95% CI 18.5–18.9%) were BIPOC. There was a significant increase in the reporting of race or ethnicity data (from 26.9% in 2000–2001 to 54.2% in 2019–2020, p&lt;0.001) and in enrollment of BIPOC (from 16.5% in 2000–2001 to 23.9% in 2019–2020, p=0.038) between 2000–2020. Trial leadership by a woman was associated with twice the adjusted odds of reporting of race or ethnicity data (OR 2.0; 95% CI 1.1–3.8; p=0.028) and an 8.4% (95% CI 1.9–15.0%; p=0.012) adjusted increase in enrollment of BIPOC. The race/ethnicity of trial leaders was not available for analysis.

Conclusions

Among HF RCTs published between 2000–2020, &lt;38% reported data on race or ethnicity, although this increased over time. Among trials reporting such data, &lt;19% of participants were BIPOC, with modest increases in enrollment over time. Trials led by women had greater adjusted odds of reporting race/ethnicity data and enrollment of BIPOC.

Funding Acknowledgement

Type of funding sources: Public grant(s) – National budget only. Main funding source(s): CIHR

Temporal trends in the inclusion of patient-reported outcomes in heart failure randomized trials published in high-impact medical journals: a systematic bibliometric review

Background

Patient reported outcomes (PROs) are important measures of health, particularly in symptomatic conditions such as heart failure (HF), and regulatory agencies have encouraged their collection in randomized controlled trials (RCTs).

Purpose

To assess temporal trends in the inclusion of patient-reported outcomes (PROs) in heart failure (HF) randomized controlled trials (RCTs) published in high-impact medical journals, explore trial characteristics associated with inclusion of PROs, and describe the quality of PRO reporting in HF RCTs.

Methods

We searched MEDLINE, EMBASE and CINAHL for studies published between January 1, 2000 and July 17, 2020. We included RCTs published in journals with impact factor ≥10. We assessed temporal trends using the Jonckheere-Terpstra test and conducted multivariable logistic regression analysis to explore trial characteristics independently associated with inclusion of PROs. We assessed the quality of PRO reporting using the Consolidated Standards for Trial Reporting PRO extension statement.

Results

We identified 12,342 articles, of which 417 RCTs met inclusion criteria. PROs were included in 224 RCTs (53.7%, 95% confidence interval [CI]: 48.8%-58.6%), of which 44 (19.6%) reported PRO as primary or co-primary endpoint. The proportion of RCTs with PROs increased significantly between 2000–2003 (37.4%) and 2016–2020 (65.1%) (p&lt;0.001). PROs had higher odds of inclusion in RCTs that were multicenter (odds ratio [OR]: 1.95; 95% CI: 1.05–3.64; p=0.036); medium-sized (n=51–250) (OR: 2.29; 95% CI: 1.24–4.23; p=0.008); coordinated in in Central and South America (OR: 6.79; 95% CI: 1.34–34.36; p=0.021); and assessed health services (OR: 4.21; 1.97–8.98; p&lt;0.001), device / surgical (OR: 6.24; 95% CI 3.05–12.80; p&lt;0.001), or exercise and rehabilitation interventions (OR: 3.98; 95% CI 1.59–9.97; p=0.003). A majority (54.9%) of the 224 RCTs reported four or less of the eleven CONSORT-PRO items. The median number of CONSORT-PRO items reported was 4 (interquartile range [IQR] 3–6 items per trial), with improved reporting in trials with PRO as primary or co-primary endpoint.

Conclusions

PROs are included in in just over half of HF RCTs, with an increased reporting of PROs between 2000 and 2020. Large, pharmacotherapy trials conducted in North America and Europe were less likely to include PROs. The quality of PRO reporting in HF RCTs was modest, with trials most often reporting four of eleven CONSORT-PRO items.

Funding Acknowledgement

Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Canadian Institutes of Health Research

Global representation of heart failure clinical trial leaders and collaborators: a systematic bibliometric review 2000–2020

Aims

Heart Failure (HF) has a disproportionate burden in low- and middle-income countries. The geographic representation of those who lead HF randomized clinical trials (RCTs) may not reflect the geographic burden of disease. We assessed temporal trends and trial characteristics associated with leadership outside Europe and North America, and explored whether there was a geographic association between trial leadership and participant enrolment.

Methods and results

We searched MEDLINE, EMBASE, and CINAHL for HF RCTs published in journals with an impact factor ≥10 between January 1, 2000, and June 17, 2020. We used the Jonckheere-Terpstra test to assess temporal trends and multivariable logistic regression models to determine associations between predictor and outcome variables.

There were 414 eligible RCTs. Only 80 of 828 trial leaders (9.7%; 95% CI: 7.8% to 11.8%), and 453 of 4656 collaborators (9.7%; 95% CI: 8.8% to 10.6%) were from regions outside Europe and North America, with no temporal change in geographic representation. The odds of trial leadership outside Europe and North America were significantly lower with industry versus public funding (OR: 0.33; 95% CI: 0.15 to 0.75; P=0.008). Trial leadership outside Europe and North America was associated with enrolment of patients outside Europe and North America (OR: 10.0; 95% CI 5.6–19.0; P&lt;0.001).

Conclusion

Trial leadership outside Europe and North America is rare, particularly in industry funded trials, and is associated with participant enrolment in regions with disproportionate disease burden. Building research capacity and networks in under-represented regions could increase generalizability of trial results.

Funding Acknowledgement

Type of funding sources: None.

Prediction of Medical Concepts in Electronic Health Records: Similar Patient Analysis

Background

Medicine 2.0—the adoption of Web 2.0 technologies such as social networks in health care—creates the need for apps that can find other patients with similar experiences and health conditions based on a patient’s electronic health record (EHR). Concurrently, there is an increasing number of longitudinal EHR data sets with rich information, which are essential to fulfill this need.

Objective

This study aimed to evaluate the hypothesis that we can leverage similar EHRs to predict possible future medical concepts (eg, disorders) from a patient’s EHR.

Methods

We represented patients’ EHRs using time-based prefixes and suffixes, where each prefix or suffix is a set of medical concepts from a medical ontology. We compared the prefixes of other patients in the collection with the state of the current patient using various interpatient distance measures. The set of similar prefixes yields a set of suffixes, which we used to determine probable future concepts for the current patient’s EHR.

Results

We evaluated our methods on the Multiparameter Intelligent Monitoring in Intensive Care II data set of patients, where we achieved precision up to 56.1% and recall up to 69.5%. For a limited set of clinically interesting concepts, specifically a set of procedures, we found that 86.9% (353/406) of the true-positives are clinically useful, that is, these procedures were actually performed later on the patient, and only 4.7% (19/406) of true-positives were completely irrelevant.

Conclusions

These initial results indicate that predicting patients’ future medical concepts is feasible. Effectively predicting medical concepts can have several applications, such as managing resources in a hospital.

FRI0423 CLINICAL BURDEN OF TREATING COMMERCIALLY-INSURED OSTEOARTHRITIS PATIENTS WITH PRESCRIPTION NSAIDS

Background:

Prescription NSAIDs/Cox-2s (“NSAIDs”) are commonly prescribed by physicians to treat patients with chronic pain, and much is known about the potential negative outcomes associated with their use1. Such negative outcomes include gastrointestinal (“GI”) issues and hepatorenal toxicity.1In addition, CV risk of Cox-2s has not been completely clarified.1,2However, less is known about the extent to which these outcomes are pervasive and problematic in specific patient populations such as those diagnosed with osteoarthritis (“OA”).

Objectives:

The goal of this research is to assess the clinical burden of commercially-insured patients diagnosed with OA of the hip and/or knee before and after treatment with prescription NSAIDs, in a large, national database in recent years.

Methods:

The Optum Healthcare Solutions, Inc. data (1/2012-3/2017) were used to identify patients ≥18 years old with ≥2 diagnoses of hip and/or knee OA, and ≥90 days supply of NSAIDs during the three-year period from first prescription (index date) after their first OA diagnosis. Patients were required to be continuously-enrolled during the six months before (baseline period) and 36 months after (follow-up period) the index date. Selected clinical outcomes such as GI issues, CV events, and renal toxicity were compared between the baseline and follow-up periods. Costs and resource use were normalized to account for differential duration in analytic time periods.

Results:

Data for 22,435 patients (60.8% as female, with an average age of 62) with hip and/or knee OA were analyzed. On average, patients were prescribed NSAIDs for 489 days during the follow-up period. From the baseline period to follow-up period, negative clinical outcomes associated with GI issues increased by 393% (1.5% v 7.5%), driven by a 667% (0.3% v 2.3%) increase in acute GI hemorrhages. Additionally, negative clinical outcomes associated with CV events increased by 73% (40.7% v 70.6%), largely due to a 600% (0.3% v 2.1%) increase in acute myocardial infarction. Lastly, negative clinical outcomes associated with renal toxicity increased by 433% (1.5% v 8.0%), with a 740% (0.5% v 4.2%) increase in acute renal failure being the most substantial.

Conclusion:

These findings suggest that prescribing of NSAIDs among OA patients is associated with an increase in negative clinical outcomes. This suggests that new treatment options other than NSAIDs should be evaluated.

References:

[1]van Laar M, et al. Pain treatment in arthritis-related pain: beyond NSAIDs.Open Rheumatol J. 2012;6:320–330.

[2]Nissen SE, et al. Cardiovascular safety of celecoxib, naproxen, or ibuprofen for arthritis.NEJM2016;2519-2529.

Disclosure of Interests:

Stuart Silverman Consultant of: Pfizer and Eli Lilly for this project., Speakers bureau: Amgen, Radius, James Rice Consultant of: Pfizer and Eli Lilly have funded this project., Alan White Consultant of: Pfizer and Eli Lilly have funded this project., Nguyen Le Consultant of: Pfizer and Eli Lilly have funded this project., Michael Somma Consultant of: Eli Lilly and Pfizer have funded this project., Craig Beck Shareholder of: Pfizer, Employee of: Pfizer, Rebecca Robinson Shareholder of: Eli Lilly, Employee of: Eli Lilly, Patricia Schepman Shareholder of: Pfizer, Employee of: Pfizer

Time Trend of Liver Cancer Among Subpopulation of Ages Less Than 40 During 2005-2014 in Nghe An, Viet Nam

Background: Liver cancer has been leading cancer mortality nationwide in Viet Nam. Aim: The aim was to examine time trends of liver cancer mortality by performing population-based mortality registration in Nghe An province, 2005-2014. Methods: We yearly collected data from all 480 Commune Health Stations from 2005 to 2014 using the designed form of “Mortality Registration” with a guideline of underlying-, immediate, - and contribute cause of death. Five variables were included name, age, sex, date of death and cause of death. Average residents of each commune were also yearly reported. All cases were coded ICD-10, then liver cancer (C22) was derived. Age-standardized rate (ASR) was estimated. Trend of liver cancer was observed by estimated mortality rates ratio and 95% confident interval (MRR) for 5-period of 2005-2006 (reference), 2007-2008, 2009-2010, 2011-2012, 2013-2014, adjusted for total mortality rate and proportion of cases with unknown causes of death, for total, subgroup of ages 40 or older, and ages less than 40. Results: Among registered 7,667 cases of liver cancer, there were 855 cases aged less than 40. Mortality rates per 100,000 (ASR) were increased from 37.4 to 52.8 in men and 9.2 to 12.6 in women, from 2005 to 2014, respectively. The increased trends were significantly seen for the subpopulation ages 40 or older only, MRR=1.56, 95% CI: 1.38-1.76 in men and MRR=1.67, 95% CI: 1.36-2.04 in women, all P trend < 0.05. In contrast, the decreased trends were significantly observed for the subpopulation ages less than 40, MRR=0.72, 95% CI: 0.54-0.97 in men and MRR=0.47, 95% CI: 0.22-0.99 in women, all P trend < 0.05. Conclusion: The significant decline risks of death from liver cancer for the subpopulation ages less than 40 during 2005-2014 might be explained by HBV vaccine nationwide from 1998 to date and reduction of environmental factors induced this cancer sites in improving healthy environments after the war ended in 1975. Key-words: Liver-cancer, population-based mortality registration, Viet Nam, HBV-vaccine.

Current Achievement and Challenges of Food, Cooking Methods and Cancer Control: A Pooled Multivariable-Adjusted Analysis for the Association Between Meat Mutagens Intake and Colorectal Adenoma and Cancer

Background: Heterocyclic amines (HCAs), a group of dietary carcinogens, have been hypothesized to increase risk of colorectal adenoma (CRA) cancer (CRC). However, previous findings have been inconsistences. Aim: The aim was to pool analysis of multivariable-adjusted relative risks from published articles for the association between HCAs intake and the risk of CRD and CRC. Methods: We used all available 34 published articles to access the multivariable-adjusted relative risk (hazard ratio or odds ratio) and their 95% confidence interval (95% CI) to estimate ln(RR) and se(ln(RR)). The HCAs 2-amino-3,8-dimethylimidazo[4,5,-f]quinoxaline (MeIQx), 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP), 2-amino-3,4,8-trimethylimidazo[4,5,-f]quinoxaline (DiMeIQx), and meat-derived mutagenicity (MDM) were examined. The random pooled RR, 95% CI was analyzed in using ln(RR) and se(ln(RR)) for separated prospective cohort studies and case-control studies by STATA-10. Results: Prospective cohort studies have examined for 5,684 CRC occurred among 528,111 registered participants; and for 2,484 CRA occurred among 136,902 registered participants. Case-control studies have examined for 12,753 CRC and 17,152 matched controls; and for 10,821 CRA and 45,099 matched controls. PhIP was majority contributed by cooked chicken (54%–74%). MeIQx was mainly contributed by red meat (83%–92%). For prospective cohort studies on CRC, fifth versus first-quintile, the random pooled RR, (95% CI) was 1.11, (1.00, 1.23), P = 0.052 for MDM; 1.00, (0.91, 1.09), P = 0.958 for PhIP; 1.12, (1.03, 1.22), P = 0.009 for MeIQx; 1.03, (0.87, 1.22), P = 0.763 for DiMeIQx. For CRA, third versus first-quintile, the random pooled RR, (95% CI) was 1.13, (0.92, 1.39), P = 0.237 for MDM; 1.11, (0.96, 1.29), P = 0.169 for PhIP; 1.14, (1.00, 1.30), P = 0.048 for MeIQx; 1.00, (0.87, 1.15), P = 0.999 for DiMeIQx. For case-control studies, the null association was observed for PhIP, DiMeIQx and CRC. A positive association was significantly seen for MeIQx, MDM and both CRC and CRA. Conclusion: Because intake of MeIQx significantly increased the risk of both CRC and CRA in both prospective cohort and case-control studies, the findings support to the hypothesis of the role of underlying mechanisms of red meat induced CRC.

A phase Ib study of the MEK inhibitor GSK1120212 combined with gemcitabine in patients with solid tumors: Interim results

278

Background: GSK1120212 (212) is a reversible, allosteric inhibitor of MEK1/MEK2. The objectives of this open-label, single-arm study are to evaluate the safety, pharmacokinetics (PK), and anti-tumor activity of 212 + gemcitabine (gem), and to determine the recommended phase II regimen (RP2R) in patients (pts) with advanced solid tumors.

Methods: 212 (1-2.5mg) is given continuously, orally, once daily. Gem (1,000mg/m2) is infused on days 1, 8, and 15 every 28 days. Doses are escalated to the maximum tolerated dose (MTD) and followed by an expansion cohort to confirm the RP2R tolerability.

Results: 28 pts received ≥ 1 dose of 212 + gem, including 8 pancreatic, 6 breast, and 4 non-small cell lung (NSCLC) cancer pts. The MTD and RP2R is 2mg 212 + 1,000mg/m2 gem. Dose-limiting toxicities (DLTs) are G3/G4 febrile neutropenia (n=2), G3 AST elevation (n=2), and G2 uveitis (n=1). 16 serious adverse events (SAEs) were reported; 5 were considered to be related to study drugs (1 pneumonitis, 3 febrile neutropenia, 1 dyspnea). All DLTs and SAEs have resolved. The most common AEs at the RP2R (n=18) were rash (78%), fatigue (67%), thrombocytopenia (61%), neutropenia (50%), decreased appetite (50%), nausea (44%), diarrhea and constipation (39%); all ≤ G2 except thrombocytopenia (17% ≥ G3) and neutropenia (33% ≥ G3). Co-administration did not affect the PK profiles of 212 or gem. 25 pts had measurable disease at baseline. 1 pancreatic cancer pt with previous radiotherapy and 2 cycles of gem achieved a partial response and stayed on study for 6 months. 3 additional pancreatic cancer pts reported stable disease; 2 of which were on the study for 3.5-5 months and the third pt continues in the study. 1 triple-negative breast cancer pt, refractory to chemotherapy, and 1 parotid cancer pt experienced a complete response of their target lesions.

Conclusions: 212 + gem is tolerable with an acceptable safety profile in this pt population, with evidence of clinical activity in pancreatic cancer. A randomized phase II study in previously untreated patients with metastatic pancreatic cancer is underway to investigate the clinical activity of this combination.

[Table: see text]

Protective Effect of Estrogen (E2) and Increased Risk of E2 Plus Progestin (Prog) on Breast Cancer (BrCa). The 2009 Review of the Women's Health Initiative (WHI) Hormone Replacement Therapy (HRT) Published Trials

INTRODUCTION & METHODOLOGY.The objective of this review is to identify from the two WHI published randomized HRT trials the BrCa risk or benefit, with a special focus on different impact of E2-alone vs E2+Prog on human breast carcinogenesis (BrCa-genesis).In the 1st trial (JAMA 2002, 2007), healthy women age 50 - 79 were randomized to E2 plus Prog versus placebo (Plac). Reviewed were Hazard ratios (HRs) for all pts and also for those with or without prior Hormone use. Mean follow up was 5.2 years.In the 2nd trial (JAMA 2004, 2006), women with hysterectomy age 50 - 79 were randomized to E2 alone (without Prog) versus Plac. Mean follow up was 7.1 years. Reviewed were all pts, and also subsets with or without prior E2 or Prog; with or without prior history of Benign Breast Disease (PHBrD); and with or without prior history of a 1st degree relative with BrCa (PH1stRelBrCa).RESULTS.Invasive BrCa incidence:N (%)HR95% C.I.1st TrialAll participants:16,608 (100%)1.260.83 - 1.97No prior Hormone use:12,304 (74.1%)1.060.81 - 1.38Prior Hormone use &lt;5 yrs:3,005 (18.1%)2.131.15 - 3.94*Prior Hormone use 5 - 10 yrs:783 (4.7%)4.611.01 - 21.02*2nd TrialAll participants:10,739 (100%)0.800.62 - 1.04No prior hormone use:5,763 (53.7%)0.650.46 - 0.92*Prior E2 use:4,976 (46.3%)0.980.67 - 1.44Prior E2 + PROG use:468 (4.4%)2.350.60 - 9.14No PHBrD:7,681 (71.5%)0.570.41 - 0.78*No PH1stRelBrCa8,554 (80.0%)0.680.50 - 0.92*CONCLUSION:1. Results from the 1st trial show that concomitant use of E2 with Prog, particularly in women with prior Hormone use, is associated with increased BrCa hazards.2. The results of the 2nd HRT trial indicate that use of HRT with E2 alone is associated with a reduction in BrCa incidence, significant in the majority of women with no prior BrCa risk factors. Also the results show that prior Prog but not prior E2 elevate the risk.3. These data indicate that HRT with E2 alone has a BrCa protective effect, while adding Prog increases the risk.4. These results are also compatible with a dual E2 effect on BrCa-genesis, where exogenous E2 are protective, but endogenous E2 [endE2] are carcinogenic, as shown from decades of anti-end E2 research with Tamoxifen, oophorectomy or Aromatase inhibitors.5. Based on these data we propose that in women with hysterectomy and no prior history of risk factors, the use of HRT with E2 alone as chemoprevention is plausible, and needs urgent verification.* significance, "p" &lt;0.05.

Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 908.

Preoperative tumor markers at diagnosis in women with malignant mixed müllerian tumors/carcinosarcoma of the uterus

CA125 is a well-recognized marker for endometrial cancer. Uterine malignant mixed müllerian tumors (MMMTs) are increasingly being recognized as an aggressive adenocarcinoma, not a sarcoma. There are no data in the literature regarding CA125 in this malignancy. One hundred twelve women with surgically staged MMMT, diagnosed between July 1990 and September 2005, had a retrospective chart review performed. Preoperative CA125 levels were available in 29 (26%) women. Seventeen (49%) women had levels above the upper limit of normal of 35 kμ/L. Mean levels increased with increasing surgical stage: stage I 53.4 kμ/L; stage II 122.5 kμ/L; stage III 147.1 kμ/L; and stage IV 428.4 kμ/L. Elevated levels of CA19-9, CEA, and CA15-3 were found in 8%, 12%, and 25%, respectively.

Myocardial protection by isoflurane vs. sevoflurane in ultra-fast-track anaesthesia for off-pump aortocoronary bypass grafting

BACKGROUND AND OBJECTIVES

Volatile anaesthetics have gained more popularity recently due to the potential for cardiac protection. Ultra-fast-track anaesthesia implies the immediate extubation after cardiac surgery. The purpose of this prospective randomized double-blind controlled study is to compare the cardioprotective effects of sevoflurane and isoflurane in off-pump cardiac bypass surgery.

METHODS

Forty patients undergoing elective off-pump cardiac bypass surgery with high thoracic epidural analgesia and immediate extubation at the end of surgery were randomized into two groups. During surgery, anaesthesia was provided with either 1 minimum alvelolar anaesthetic concentration of sevoflurane or 1 minimum alvelolar anaesthetic concentration of isoflurane. Troponin-T, creatine kinase-MB, left ventricular wall motion anomalies, time to extubation, respiratory functions and haemodynamic parameters were compared between the two groups by analysis of variance.

RESULTS

All patients were successfully extubated in the operating theatre with minimal postoperative pain. Serial creatine kinase-MB and troponin-T concentrations were not significantly different between the two volatile agents. Haemodynamic stability throughout surgery and contractility was not different between groups. However, extubation time was significantly shorter with sevoflurane (10 +/- 5 min) compared to isoflurane (18 +/- 4 min).

CONCLUSION

This study indicates that during off-pump cardiac bypass surgery, sevoflurane and isoflurane provide the same ischaemic cardioprotective effects. There is no difference for heart contractility and haemodynamic values during and after off-pump cardiac bypass surgery between the two agents. Sevoflurane allows a more rapid recovery from anaesthesia, but this does not translate into better pulmonary function or haemodynamics. Both agents are similar in ultra-fast-track off-pump cardiac bypass surgery.

Pulsed high-intensity focused ultrasound enhances uptake of radiolabeled monoclonal antibody to human epidermoid tumor in nude mice

The aim of this study was to determine if pulsed high-intensity focused ultrasound (HIFU) exposures could enhance tumor uptake of (111)In-MX-B3, a murine IgG1kappa monoclonal antibody directed against the Le(y) antigen.

METHODS

MX-B3 was labeled with (111)In, purified, and confirmed for its binding to the antigen-positive A431 cell line. Groups of nude mice were inoculated subcutaneously with A431 tumor cells on both hind flanks. A tumor on one flank was treated with pulsed-HIFU; the other tumor was used as an untreated control. Within 10 min after the HIFU exposure, the mice received intravenous (111)In-MX-B3 for imaging and biodistribution studies. Mice were euthanized at 1, 24, 48, and 120 h after injection for biodistribution studies.

RESULTS

The HIFU exposure shortened the peak tumor uptake time (24 vs. 48 h for the control) and increased the peak tumor uptake value (38 vs. 25 %ID/g [percentage injected dose per gram] for the control). The HIFU effect on enhancing tumor uptake was greater at earlier times up to 24 h, but the effect was gradually diminished thereafter. The HIFU effect on enhancing tumor uptake was substantiated by nuclear imaging studies. HIFU also increased the uptake of the antibody in surrounding tissues, but the net increase was marginal compared with the increase in tumor uptake.

CONCLUSION

This study demonstrates that pulsed-HIFU significantly enhances the delivery of (111)In-MX-B3 in human epidermoid tumors xenografted in nude mice. The results of this pilot study warrant further evaluation of other treatment regimens, such as repeated HIFU exposures for greater delivery enhancement of antibodies labeled with cytotoxic radioisotopes or pulsed-HIFU exposure in addition to a combined therapy of (90)Y-B3 and taxol to enhance the synergistic effect.

Synergistic antitumor activity of taxol and immunotoxin SS1P in tumor-bearing mice

PURPOSE

To investigate the combined antitumor activity in mice of immunotoxin SS1P and Taxol.

METHODS

Immunodeficient mice were implanted with A431/K5 tumors expressing mesothelin. Established tumors were treated i.v. with immunotoxin SS1P alone, i.p. with Taxol alone, or with the two agents together. SS1P was radiolabeled with (111)In and used to study the effect of Taxol on its uptake by A431/K5 tumors.

RESULTS

Using doses at which either agent alone caused stabilization of tumor growth, the combination was synergistic causing long-lasting complete remissions in many animals. In contrast, synergy was not observed when the same cells were treated with these agents in vitro. Tumor uptake of (111)In-SS1P was not affected by treatment with Taxol.

CONCLUSION

The combination of Taxol and SS1P exerts a synergistic antitumor effect in animals but not in cell culture. This effect is not secondary to increased tumor uptake of the immunotoxin. Synergy could be due to improved immunotoxin distribution within the tumor or could involve factors released by other cell types in the tumors.

Multimodal nanoprobes for radionuclide and five-color near-infrared optical lymphatic imaging

Current contrast agents generally have one function and can only be imaged in monochrome; therefore, the majority of imaging methods can only impart uniparametric information. A single nanoparticle has the potential to be loaded with multiple payloads. Such multimodality probes have the ability to be imaged by more than one imaging technique, which could compensate for the weakness or even combine the advantages of each individual modality. Furthermore, optical imaging using different optical probes enables us to achieve multicolor in vivo imaging, wherein multiple parameters can be read from a single image. To allow differentiation of multiple optical signals in vivo, each probe should have a close but different near-infrared emission. To this end, we synthesized nanoprobes with multimodal and multicolor potential, which employed a polyamidoamine dendrimer platform linked to both radionuclides and optical probes, permitting dual-modality scintigraphic and five-color near-infrared optical lymphatic imaging using a multiple-excitation spectrally resolved fluorescence imaging technique.

Radiolabeled high affinity peptidomimetic antagonist selectively targets alpha(v)beta(3) receptor-positive tumor in mice

OBJECTIVES

The aim of this research was to synthesize radiolabeled peptidomimetic integrin alpha(v)beta(3) antagonists that selectively target integrin alpha(v)beta(3) receptor and clear rapidly from the whole body.

METHODS

Integrin alpha(v)beta(3) antagonists, 4-[2-(3,4,5,6-tetrahydropyrimidine-2-ylamino)ethyloxy]benzoyl-2-(S)-aminoethylsulfonyl-amino-beta-alanine (IA) and 4-[2-(3,4,5,6-tetrahydro-pyrimidin-2-ylamino)-ethyloxy]benzoyl-2-(S)-[N-(3-amino-neopenta-1-carbamyl)]-aminoethylsulfonylamino-beta-alanine hydrochloride (IAC), a hydrophobic carbamate derivative of IA, were conjugated with 2-p-isothiocyanatobenzyl-DOTA at the amino terminus and labeled with (111)In. The (111)In labeled IA and IAC were subjected to in vitro receptor binding, biodistribution and imaging studies using nude mice bearing the receptor-positive M21 human melanoma xenografts.

RESULTS

The (111)In-labeled IA (40%) and -IAC (72%) specifically bound in vitro to alpha(v)beta(3) (0.8 microM) at a molar excess. This receptor binding was completely blocked by a molar excess of cold IA to alpha(v)beta(3). The higher receptor-binding affinity of the (111)In-labeled IAC was reflected in higher tumor uptake and retention: 5.6+/-1.4 and 4.5+/-0.7 %ID/g vs. 3.8+/-0.9 and 2.0+/-0.3 %ID/g for the (111)In-labeled IA at 0.33 and 2 h. The tumor uptakes were inhibited by the co-injection of 200 microg of IA, indicating that the uptake was receptor mediated. These antagonists were excreted primarily via the renal system. The (111)In activity retained in the whole body was quite comparable between the (111)In-labeled IA (24% ID) and the (111)In-labeled IAC (33% ID) at 2 h. The higher peak tumor uptake and longer retention resulted in higher tumor-to-background ratios for the (111)In-labeled IAC at 2 h with 9.7, 2.3, 0.8, 1.9, 7.1, 2.2, 0.9, 3.7 and 9.9 for blood, liver, kidney, lung, heart, stomach, intestine, bone and muscle, respectively. The imaging studies with the (111)In-labeled IAC also clearly visualized the receptor-positive tumor at 4 h.

CONCLUSIONS

The (111)In-labeled IAC showed an improve tumor targeting kinetics with rapid accumulation and prolonged retention in the alpha(v)beta(3) receptor-positive tumor. This together with the rapid whole-body clearance pharmacokinetics warrants further studies on this IAC analog for molecular imaging of tumor-induced angiogenic vessels and various malignant human tumors expressing the receptor.

Use of antibody as carrier of oligomers of peptidomimetic alphavbeta3 antagonist to target tumor-induced neovasculature

Sulfhydryl selective reactions were explored to conjugate oligomers of a peptidomimetic integrin alphavbeta3 antagonist, 4-[2-(3,4,5,6-tetrahydropyrimidine-2-ylamino)ethyloxy]benzoyl-2-(S)-aminoethylsulfonylamino-beta-alanine (IA) to monoclonal antibody (MoAb) to increase integrin alphavbeta3 receptor-binding avidity. To generate sulfhydryl groups, N-succinimidyl-S-acetylthioacetate (SATA) was conjugated to both MoAb and IA. Sulfhydryl groups were then generated upon the deacetylation of the protecting acetyl group from the S-acetylthioacetato (ATA) moiety of MoAb-(ATA)n or IA-ATA with 0.02 M hydroxylamine in the presence of 1 mM EDTA at pH 7.2. The major focus was on optimizing the reaction concentrations, molar ratios, and reaction pH to conjugate high levels of IA-(A-SH) to MoAb-(A-SH)n without causing the inter- and intramolecular cross-linking of MoAb. Stepwise reactions of MoAb-(A-SH)n (15 microM MoAb) with a homobifunctional cross-linker, 1,8-bis(maleimido)diethylene glycol (BM[PEO]2) at a >50x molar excess to the -SH, followed by the reaction of the purified product MoAb-(A-S-succinimidomaleimido-[PEO]2)n with IA-(A-SH) at pH 7.2 afforded monomeric MoAb-(A-S-succinimido-[PEO]2-succinimido-S-A-IA)n with <10% high molecular weight oligomeric MoAb. Monomeric MoAb-(A-S-S-[PEO]2-S-S-A-IA)10 (MoAb-IA10) radiolabeled with 111In using 2-(p-isothiocyanatobenzyl)cyclohexyl-DTPA and with 125I using the Iodogen method showed >70% bindability to 0.4 microM alphavbeta3. When injected iv to nude mice with the receptor-positive M21 tumor, MoAb-IA10 radiolabeled with both 111In and 125I accumulated rapidly and was retained in the tumor for a 44 h period while the radioactivity cleared rapidly from the blood, thereby resulting in increasing tumor-to-blood ratios over time. The tumor uptake was similar between the 125I label and the 111In label for a 44 h period. In contrast, the blood radioactivity was lower, but liver and other organ uptakes were much higher for the 111In label than for the 125I. The 111In label produced higher tumor-to-blood ratios but much lower tumor-to-organ ratios than the 125I. The rapid blood clearance, a short peak tumor uptake time, and a low peak tumor uptake value with prolonged tumor retention of this macromolecule appear to support a hypothesis that MoAb-IA10 primarily binds to alphavbeta3 receptors on angiogenic vessels, but not on the tumor. This hypothesis was substantiated by the fluorescence microscopic analysis of FITC-MoAb-IA10, which showed that FITC-MoAb-IA10 outlined neovasculatures but not tumor cells at 4 and 21 h ex vivo. Additional proof was observed when blood vessels outlined with rhodamine-lectin, which specifically binds to blood vessels, were superimposable on neovasculatures outlined with FITC-MoAb-IA10.

Comparison of three different epidural solutions in off-pump cardiac surgery: pilot study

BACKGROUND

Immediate extubation using thoracic epidural analgesia (TEA) has become more popular after off-pump coronary artery bypass grafting (OPCAB). In this randomized prospective double-blind study, we present the first comparison of preoperative and postoperative haemodynamics during different regimens of TEA for immediate extubation after cardiac surgery.

METHODS

Sixty patients undergoing OPCAB were enrolled in this study. TEA was installed >1 h before application of heparin at levels T2-T4. Analgesia was provided by bupivacaine 0.25%, 8 ml, 15 min before surgery and extubation, and at 10 ml h(-1) during surgery and up to 72 h afterwards using one of the following regimens: bupivacaine 0.125% alone, bupivacaine 0.125% with fentanyl 3 microg ml(-1) or bupivacaine 0.125% with clonidine 0.6 microg ml(-1). Patients were block-randomized for one of the three treatments. Pain scores and infusion rates of TEA were assessed up to 48 h after surgery. Respiratory function was assessed by Pa(o2) and Pa(o2) immediately after surgery, and haemodynamic stability was recorded in the form of heart rate and diastolic and systolic blood pressure.

RESULTS

Patient characteristics, respiratory function and haemodynamic stability did not vary between the three groups. Pain control was very good and was not significantly different between the groups using similar infusion rates after surgery. Paraesthesia in dermatomes T1 or C8 occurred equally in all three groups. There was no neurological complication related to TEA in this study.

CONCLUSIONS

We conclude that immediate extubation after OPCAB using TEA is feasible with different TEA regimens. Respiratory function, haemodynamic stability and pain control are not different between TEA with bupivacaine alone, bupivacaine with fentanyl or bupivacaine with clonidine.

Hodgkin’s lymphoma in Chinese migrants to British Columbia:a 25-year survey

BACKGROUND

Compared with the West, Hodgkin's lymphoma in Oriental countries is characterized by a lower incidence rate and a higher proportion of mixed cellularity histology. Both environmental and genetic factors may be involved.

PATIENTS AND METHODS

The incidence and pattern of pathology of Hodgkin's lymphoma in the migrant Chinese population (0.4 million) in British Columbia (population 3.2 million) were studied. From a computerized database, all Hodgkin's lymphoma cases diagnosed in British Columbia from 1970 to 1997 were identified. Chinese descent was determined using patient surname by standard methodology and verified from the treatment record or by patient interview. The corresponding figures from the Chinese population in Hong Kong were used for comparison. For incidence rates, the age-specific incidence of Hodgkin's lymphoma in Hong Kong was obtained from the government cancer registry. For comparison of histology subtypes, 200 Hodgkin's lymphoma records from a Hong Kong regional referral center for the same time period were reviewed. Crude and age-standardized incidence rates were calculated by 5-year intervals in terms of age and calendar year, and relative rates were compared between the three populations.

RESULTS

From 1970 to 1997, Hodgkin's lymphoma was diagnosed in 34 Chinese patients in BC, with 24 cases diagnosed from 1970 to 1994. Thus, the crude and age-adjusted incidence rates from 1970 to 1994 were 0.91 and 1.14 per 100,000 per year in the British Columbia Chinese migrant population. Within the same period, 1862 cases of Hodgkin's lymphoma were diagnosed in British Columbia, giving a provincial background crude and age-adjusted incidence rates of 5.2 and 4.87 per 100,000 per year. The number of cases in the Hong Kong Chinese population (1970-1994) was 404, giving crude and age-adjusted incidence rates of 0.32 and 0.31 per 100,000 per year, respectively. Corrected for age and calendar year trends, the observed 25-year incidence of Hodgkin's lymphoma in British Columbia Chinese was significantly lower than expected from the British Columbia background population [24 observed versus 71 expected cases; standardized incidence ratio (SIR) = 0.34; 90% confidence interval (CI) 0.24-0.48; P <0.0001]. On the other hand, it is higher than that expected by extrapolating from the Hong Kong Chinese population (24 observed versus 8.5 expected cases; SIR = 2.81; 90% CI 1.94-3.95; P <0.0001). The difference is mainly accounted for by young patients with nodular sclerosis type disease in the migrant population.

CONCLUSIONS

Although any conclusion about the impact of migration on Hodgkin's lymphoma incidence and types in the Chinese population must be considered tentative due to the small number of observed cases and confounding variables such as age, changing diagnostic standards and secular trends in Hodgkin's lymphoma rates, our data demonstrate a tendency for the Chinese population of British Columbia to take on a Western pattern of Hodgkin's lymphoma. This observation provides additional evidence that both genetic and environmental influences play a role in the pathogenesis of this lymphoma, and that environmental factors can exert their influence over a relatively short period of time.

Rapamycin causes down-regulation of CCR5 and accumulation of anti-HIV beta-chemokines: an approach to suppress R5 strains of HIV-1

Propagation of R5 strains of HIV-1 on CD4 lymphocytes and macrophages requires expression of the CCR5 coreceptor on the cell surface. Individuals lacking CCR5 (CCR5 Delta 32 homozygous genotype) are phenotypically normal and resistant to infection with HIV-1. CCR5 expression on lymphocytes depends on signaling through the IL-2 receptor. By FACS analysis we demonstrate that rapamycin (RAPA), a drug that disrupts IL-2 receptor signaling, reduces CCR5 surface expression on T cells at concentrations as low as 1 nM. In addition, lower concentrations of RAPA (0.01 nM) were sufficient to reduce CCR5 surface expression on maturing monocytes. PCR analysis on peripheral blood mononuclear cells (PBMCs) showed that RAPA interfered with CCR5 expression at the transcriptional level. Reduced expression of CCR5 on PBMCs cultured in the presence of RAPA was associated with increased extracellular levels of macrophage inflammatory protein (MIP)-1 alpha and MIP-1 beta. In infectivity assays, RAPA suppressed the replication of R5 strains of HIV-1 both in PBMC and macrophage cultures. In total PBMC cultures, RAPA-mediated inhibition of CCR5-using strains of HIV-1 occurred at 0.01 nM, a concentration of drug that is approximately 103 times lower than therapeutic through levels of drug in renal transplant recipients. In addition, RAPA enhanced the antiviral activity of the CCR5 antagonist TAK-779. These results suggest that low concentrations of RAPA may have a role in both the treatment and prevention of HIV-1 infection.

Lowering of pI by acylation improves the renal uptake of 99mTc-labeled anti-Tac dsFv: effect of different acylating reagents

Anti-Tac disulfide-stabilized variable region fragment (dsFv) was labeled with 99mTc by a preformed chelate approach using 99mTc-MAG3-trifluorophenyl (TFP) ester. Simultaneously it was acylated with TFP-lactate or succinic anhydride to decrease the isoelectric point of dsFv (pI 10). Acylation of dsFv (0.04 mM) with the lactate at a 73 times molar excess reduced the pI to 5.0-6.7, whereas acylation with succinic anhydride at a 30 times molar excess reduced the pI to 4.9-8.7. Comparative biodistribution studies performed in mice (n = 5) showed the reduced renal accumulation of the 99mTc proportional to the pI reduction. The effect of the pI on the reduced renal uptake was especially pronounced at 15 min postinjection. The reduced renal uptake was also reflected in the reduced whole-body retention, indicating that lowering the pI inhibited the tubular reabsorption of the labeled dsFv.

The HIV protease inhibitor indinavir decreases insulin- and contraction-stimulated glucose transport in skeletal muscle

In many patients with human immunodeficiency virus (HIV) treated with HIV protease inhibitors, a complication develops that resembles abdominal obesity syndrome, with insulin resistance and glucose intolerance that, in some cases, progresses to diabetes. In this study, we tested the hypothesis that indinavir, an HIV-protease inhibitor, directly induces insulin resistance of glucose transport in skeletal muscle. Rat epitrochlearis muscles were incubated with a maximally effective insulin concentration (12 nmol/l) and 0, 1, 5, 20, or 40 micromol/l indinavir for 4 h. In control muscles, insulin increased 3-O-[(3)H]methyl-D-glucose (3MG) transport from 0.15 +/- 0.03 to 1.10 +/- 0.05 micromol. ml(-)(1). 10 min(-)(1). Incubation of muscles with 5 micromol/l indinavir reduced the insulin-stimulated increase in 3MG transport by 40%, whereas 20 micromol/l indinavir reduced the insulin-stimulated increase in 3MG transport by 58%. Indinavir induced a similar reduction in maximally insulin-stimulated 3MG transport in the soleus muscle. The increase in glucose transport activity induced by stimulating epitrochlearis muscles to contract was also markedly reduced by indinavir. The insulin-stimulated increase in cell-surface GLUT4, assessed using the 2-N-4-(1-azi-2,2,2-trifluoroethyl)benzoyl-1,3-bis-[2-(3)H] (D-mannose-4-yloxy)-2-propylamine exofacial photolabeling technique, was reduced by approximately 70% in the presence of 20 micromol/l indinavir. Insulin stimulation of phosphatidylinositol 3-kinase activity and phosphorylation of protein kinase B were not decreased by indinavir. These results provide evidence that indinavir inhibits the translocation or intrinsic activity of GLUT4 rather than insulin signaling.

EGR2 mutations in inherited neuropathies dominant-negatively inhibit myelin gene expression

The identification of EGR2 mutations in patients with neuropathies and the phenotype Egr2/Krox20(-/-) have demonstrated that the Egr2 transcription factor is critical for peripheral nerve myelination. However, the mechanism by which these mutations cause disease remains unclear, as most patients present with disease in the heterozygous state, whereas Egr2(+/-) mice are phenotypically normal. To understand the effect of aberrant Egr2 activity on Schwann cell gene expression, we performed microarray expression profiling to identify genes regulated by Egr2 in Schwann cells. These include genes encoding myelin proteins and enzymes required for synthesis of normal myelin lipids. Using these newly identified targets, we have shown that neuropathy-associated EGR2 mutants dominant-negatively inhibit wild-type Egr2-mediated expression of essential myelin genes to levels sufficiently low to result in the abnormal myelination observed in these patients.

Differential human immunodeficiency virus-suppressive activity of reverse transcription inhibitors in resting and activated peripheral blood lymphocytes: implications for therapy

OBJECTIVES

Because recent evidence indicates that human immunodeficiency virus type 1 (HIV-1) propagates in resting T lymphocytes in vivo, we wanted to evaluate the antiviral effects exerted by currently used nucleoside (NRTI) and non-nucleoside analog reverse transcription inhibitors in resting lymphocytes, and compare those effects to the ones obtained in activated lymphocytes.

METHODS

Tissue culture antiviral assays in which target cells are lymphocytes present in a resting or activated state. Virus replication was measured by a reverse transcription (RT) assay. Cell viability was evaluated using a commercial 3-(4k5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay.

RESULTS

In vitro results obtained with concentrations of zidovudine and stavudine equivalent to drug levels found in plasma, showed more than 99% HIV-1 inhibition in activated lymphocytes but less than 50% virus inhibition in resting lymphocytes. Conversely, plasma levels of didanosine-inhibited HIV-1 by approximately 50% and 98% in activated and resting lymphocytes, respectively. Plasma level concentration of zalcitabine, lamivudine, and abacavir inhibited viral replication by more than 90% in both resting and activated cells.

CONCLUSIONS

These data demonstrate that specific NRTI antiretroviral agents have different activity against HIV RT, depending on the state of cell cycle of the infected cell. We suggest that the replication of HIV-1 in resting lymphocytes should be taken into account in the design of future clinical trials, as well as treatment antiretroviral regimens. Selection of combination RTIs so that they provide antiretroviral activity in both resting and activated lymphocytes may be a way to minimize treatment failure and the emergence of drug-resistant variants.

The correlation between epidermal growth factor levels in saliva and the severity of oral mucositis during oropharyngeal radiation therapy

BACKGROUND

Epidermal growth factor (EGF) is present in biologic fluids, including saliva, and plays a role in maintenance of the epithelial barrier and in healing of damaged mucosa. The purpose of this study was to assess the relation between salivary EGF and the severity of oral mucositis in patients with carcinoma of the head and neck during radiation therapy.

METHODS

Whole resting saliva (WRS) and whole stimulated saliva (WSS) were collected prior to radiation and each week during radiation treatment for 11 men and 7 women. Oral mucositis was evaluated using the National Cancer Institute (NCI) scale of 0-4 and the Oral Mucositis Assessment Scale (OMAS), which evaluates the extent of erythema (scale of 0-2) and ulcerations (scale of 0-3) in nine oral sites. The overall OMAS score of 0-45 reflected the mucosal condition. EGF was assayed in the saliva specimens.

RESULTS

The total mean radiation dose delivered to the head and neck was 5667 centigrays (cGy) in a mean of 24 fractions. Ulcerative oral mucositis occurred in 94% of patients. The mean OMAS score ranged from 2.83 in the first week of treatment to 14.77 in the fifth week. The mean WRS and WSS volumes decreased significantly from pretreatment to the first week of radiation treatment and then remained stable. A similar pattern was seen for the mean total output of EGF. A significant and negative correlation was found between higher levels of EGF in stimulated saliva and low OMAS score, reflecting less severe erythema and ulceration. A general trend showing that less tissue damage was associated with a higher EGF level in resting saliva also was illustrated. EGF levels were correlated with the OMAS score; however, no correlation was found when assessing the NCI score, which combines tissue damage with function and symptoms in a single score.

CONCLUSIONS

Radiation-induced mucositis appeared to be modified by saliva volume, total EGF, and concentration of EGF in the oral environment. Saliva volume and total EGF output decreased significantly in the first weeks of treatment and remained reduced throughout radiation therapy. The findings suggest that higher levels of EGF in saliva, particularly in stimulated saliva, prior to and during radiation treatment may be associated with less severe mucosal damage due to radiation therapy. It is also postulated that human EGF may affect the development and healing of radiation-damaged mucosa.

Primary structure of Noetia ponderosa hemoglobins: functional correlates

Homo- and heterodimeric hemoglobins have been isolated from the red cells of the arcid clam Noetia ponderosa (Np). These hemoglobins bind oxygen cooperatively. An extensively studied dimeric hemoglobin from another arcid clam, Scapaharca inaequivalvis, exhibits a molecular mechanism for cooperative ligand binding that is radically different from tetrameric vertebrate hemoglobins. In this study, the two chains found in both Noetia hemoglobins are sequenced and compared to the hemoglobins of the related clam S. inaequivalvis to determine whether Noetia hemoglobins have the structural basis for the same unusual mechanism for cooperative ligand binding and to inquire about the structural basis of absence of tetramers. Although the Noetia sequences are homologous to the Scapharca sequences, critical differences exist. The lack of tetramerization of Np subunits is most likely related to the absence of critical residues in the A and G helices that stabilize the interdimer contact seen in the Scapharca Hb tetramer. The lower affinity of the homodimer (Np-I), but particularly the heterodimer (Np-II) with respect to the homodimer and heterotetramer of Scapharca, can be due to (i) changes in the proximal heme environment and (ii) changes in the dimer interface. Interactions between Asn 100 and the heme of the other subunit are altered in Np-II due to the substitution of this residue by methionine, possibly causing the reduced O(2) affinity of the heterodimer of Noetia. (iii) Sequence changes in the E and F helices present in Np-I and Np-II could also contribute to the effect through interfacial changes. In particular, the substitution of Val for Thr in position 72 is expected to have a substantial influence on the interface. We conclude that Np dimers have the structural basis for a direct heme-heme interaction mechanism for cooperativity, as in Scapharca, but there are enough sequence changes to suggest that the pathway of interaction might be somewhat different.

DNA cleavage by 111In-labeled oligodeoxyribonucleotides

We studied the fine structure of DNA damage produced by the decay of 111In incorporated into duplex and triplex DNA strands to evaluate the usefulness of this radionuclide for sequence-specific DNA cleavage.

METHODS

Oligodeoxyribonucleotides (ODNs) were prepared with 111In attached by diethylenetriaminepentaacetic acid (DTPA) at the 5' end or 3' end through a long chemical linker or to an internal nucleotide position through a short linker. Subsequent formation of DNA duplexes and triplexes was confirmed by gel electrophoresis. The 111In-induced breaks were assayed in denaturing polyacrylamide gel electrophoresis with a single-nucleotide resolution.

RESULTS

111In-labeled oligonucleotides of high specific activity (740-1554 TBq/mmol) were synthesized. The presence of the bulky 111In-DTPA group did not impede duplex or triplex formation. Localized DNA breaks were observed in all duplexes and triplexes formed. The majority of DNA breaks in duplex formations were located within +/- 10 nucleotides from the site of attachment of the 111In-bearing linker. The yield of DNA breaks per decay was 0.38 in a duplex with internally modified ODNs. This is nearly 2 times less than the yield of DNA breaks in the same duplex with 1251 attached through the same linker. The yield of DNA breaks in the pyrimidine and purine strands of DNA triplexes with 111In attached to the triplex-forming ODNs through the linkers of different length varied from 0.05 to 0.10. The distribution of DNA breaks was wider in comparison with the duplex experiment. The lower yields of breaks per 111In decay compared with 125I may be not only the result of lower deposited energy but also of the ionic repulsion of the negatively charged 111In-DTPA group from the DNA strands.

CONCLUSION

We have shown that decay of 111In produces highly localized DNA breaks. 111In introduced into triplex- and duplex-forming ODNs through hydrocarbon linkers produces sequence-specific DNA strand breaks with an efficiency nearly comparable with that of 1251. These findings are supportive of our proposed use of 111In-ODNs for gene-specific radiotherapy.

Pharmacokinetics of 111In- and 125I-labeled antiTac single-chain Fv recombinant immunotoxin

The use of immunotoxins for cancer therapy is an attractive strategy that exploits the targeting specificity of monoclonal antibodies and their fragments as well as the exquisite toxicity of the toxins. However, few studies of immunotoxins have evaluated their biodistribution in vivo. Previous studies have used 125I for tracing immunotoxin biodistribution in mice. Because the immunotoxin works only when it is internalized and because of known problems with quick dehalogenation after internalization of antibodies, we decided to use 111In, which has greater intracellular retention than iodine.

METHODS

To trace the in vivo pharmacokinetics of the immunotoxin in mice, we labeled the antiTac(Fv)-PE38 with 111ln and compared it with 125I-labeled antiTac(Fv)-PE38. We successfully labeled antiTac(Fv)-PE38 with 111In at up to 2.96 GBq/mg. A 3- to 4-fold decrease in cytotoxicity was observed for both radiolabeled preparations. We evaluated the internalization of 111In- and 125I-labeled antiTac(Fv)PE38 into ATAC4 cells (Tac-positive) as well as their biodistribution and pharmacokinetics in vivo in mice. In addition, some mice receiving these reagents were co-infused with 30 mg L-lysine to inhibit renal accumulation.

RESULTS

Significantly more 111In- than 125I-labeled antiTac(Fv)-PE38 accumulated in the ATAC4 cells (20% versus 5% of initial surface-bound radioactivity; P < 0.001). In vivo, significantly more 111In- than 125I-labeled antiTac(Fv)-PE38 accumulated in the kidney (119 versus 31 percentage injected dose per gram [%ID/g]; P < 0.001). The tumor accumulation of 111In-labeled antiTac(Fv)-PE38 at 96 h was 13-fold greater than that of 125I-labeled antiTac(Fv)-PE38 (1.4 versus 0.1 %ID/g; P < 0.001). No antiTac(Fv)-PE38 was excreted into the urine in its intact form unless lysine was co-infused. Co-injected lysine reduced the renal accumulation of 111In-labeled antiTac(Fv)-PE38 by 62%.

CONCLUSION

We evaluated the biodistribution, pharmacokinetics, and catabolism of 111In-labeled antiTac(Fv)-PE38 and found that it differed from 125I-labeled antiTac(Fv)PE38. These studies suggest that 111In-labeled antiTac(Fv)-PE38 can be used to trace the fate of antiTac(Fv)-PE38 in humans.

Effects of insulinlike growth factor binding proteins on insulinlike growth factor-I biodistribution in tumor-bearing nude mice

This study evaluated the biodistribution and tumor targeting ability of radiolabeled insulinlike growth factor (IGF)-I. Because IGF binding proteins (IGFBPs) play a critical role in modulating IGF activity, the binding properties of 125I-labeled IGF-I to IGFBPs were investigated in vitro and in vivo. Because a large amount of the IGF-I was catabolized in vivo, we also studied the catabolism of IGF-I by tumor cells in vitro.

METHODS

125I-labeled-IGF-I was prepared using the chloramine T method. The biodistribution of 125I-labeled-IGF-I in tumor-bearing nude mice was compared between groups injected with 125I-labeled IGF-I alone or coinjected with unlabeled peptide. In vitro and in vivo chromatography studies were performed to evaluate the binding profile to IGFBPs and the degree of catabolites in serum as well as urine.

RESULTS

Data indicated that the binding of radiolabeled IGF-I to IGFBPs in vitro was dose dependent. However, there was a difference in complex formation between the serum and the heparinized plasma. In heparinized plasma, the radioactivity shifted from a 30- to 50-kDa complex to a 150-kDa complex and to a free ligand, because the binding of heparin with IGFBPs decreased its affinity for IGF-I. In plasma prepared with acid citrate dextrose a binding pattern identical to that of serum was observed. Moreover, there was a binding difference between mouse and rat. The 125I-labeled IGF-I catabolized very quickly when incubated at 37 degrees C but not at all at 4 degrees C. In tumor-bearing nude mice, the uptake of radioactivity in normal tissues decreased quickly, particularly in the kidneys. In mice coinjected with unlabeled carrier, the radioactivity in most normal tissues was lower and the tumor uptake higher than in the mice without carrier.

CONCLUSION

These data confirm that 125I-labeled IGF-I is avidly bound to IGFBPs, both in vitro and in vivo. By partially saturating this binding with unlabeled peptides, a favorable biodistribution was achieved, including faster clearance from normal tissue and higher tumor uptake, which resulted in better tumor-to-nontumor ratios. Nevertheless, the rapid catabolism and release of the radiolabel from tumor tissue result in a suboptimal targeting agent.