Neural network approximation of optimal controls for stochastic reaction-diffusion equations

We present a numerical algorithm that allows the approximation of optimal controls for stochastic reaction-diffusion equations with additive noise by first reducing the problem to controls of feedback form and then approximating the feedback function using finitely based approximations. Using structural assumptions on the finitely based approximations, rates for the approximation error of the cost can be obtained. Our algorithm significantly reduces the computational complexity of finding controls with asymptotically optimal cost. Numerical experiments using artificial neural networks as well as radial basis function networks illustrate the performance of our algorithm. Our approach can also be applied to stochastic control problems for high dimensional stochastic differential equations and more general stochastic partial differential equations.

Craniectomy size and decompression of the temporal base using the altered posterior question-mark incision for decompressive hemicraniectomy

The altered posterior question-mark incision for decompressive hemicraniectomy (DHC) was proposed to reduce the risk of intraoperative injury of the superficial temporal artery (STA) and demonstrated a reduced rate of wound-healing disorders after cranioplasty. However, decompression size during DHC is essential and it remains unclear if the new incision type allows for an equally effective decompression. Therefore, this study evaluated the efficacy of the altered posterior question-mark incision for craniectomy size and decompression of the temporal base and assessed intraoperative complications compared to a modified standard reversed question-mark incision. The authors retrospectively identified 69 patients who underwent DHC from 2019 to 2022. Decompression and preservation of the STA was assessed on postoperative CT scans and CT or MR angiography. Forty-two patients underwent DHC with the standard reversed and 27 patients with the altered posterior question-mark incision. The distance of the margin of the craniectomy to the temporal base was 6.9 mm in the modified standard reversed and 7.2 mm in the altered posterior question-mark group (p = 0.77). There was no difference between the craniectomy sizes of 158.8 mm and 158.2 mm, respectively (p = 0.45), and there was no difference in the rate of accidental opening of the mastoid air cells. In both groups, no transverse/sigmoid sinus was injured. Twenty-four out of 42 patients in the modified standard and 22/27 patients in the altered posterior question-mark group had a postoperative angiography, and the STA was preserved in all cases in both groups. Twelve (29%) and 5 (19%) patients underwent revision due to wound-healing disorders after DHC, respectively (p = 0.34). There was no difference in duration of surgery. Thus, the altered posterior question-mark incision demonstrated technically equivalent and allows for an equally effective craniectomy size and decompression of the temporal base without increasing risks of intraoperative complications. Previously described reduction in wound-healing complications and cranioplasty failures needs to be confirmed in prospective studies to demonstrate the superiority of the altered posterior question-mark incision.

Abstract P6-08-06: Neoadjuvant tamoxifen therapy synchronizes ERα binding and gene expression profiles

Background: The majority of breast cancer patients are diagnosed with ERα-positive breast cancer. Most ERα-positive patients are treated with adjuvant endocrine therapy — typically tamoxifen or aromatase inhibitors — to block cellular proliferation. Although these treatments are considered successful, resistance is common. Notably, cross-resistance between the two types of therapies is not always observed suggesting molecular heterogeneity and underlining the need for development of personalized treatments. The Anastrozole, Fulvestrant or Tamoxifen Exposure — Response in molecular profile study (AFTER study, NCT00738777) aims to investigate prospectively whether short-term treatment can induce molecular changes indicative of pre-operative therapy response. Study Design: ERα-positive breast cancer patients are included in this open-label multicenter study. Post-menopausal patients are randomized between tamoxifen, anastrozole and fulvestrant and pre-menopausal and male patients receive tamoxifen. Treatment occurs during the pre-operative window between diagnosis and surgery (4±2 weeks). Clinical characteristics collected are ERα/PR and HER2 status as well as lymph-node status. The primary endpoint is the decrease in tumor cell proliferation, as assessed by Ki67 gene expression and published cell proliferation gene expression signatures. All data are collected from both pre- and post-treatment samples. Additionally, we will compare the changes induced by treatment in gene expression, ERα/DNA binding interactions, DNA copy number, endoxifen and estradiol levels. Results: Among 67 patients currently enrolled, we examined the data from the subset of 28 tamoxifen treated patients. ERα and PR levels did not differ significantly between pre- and post-treatment. All tumors were HER2-negative. Proliferation examined by Ki67 (IHC and gene expression, MKI67) was significantly lower in post-treatment samples (P < 0.01). A significant association was identified with the change in gene expression proliferation signature score and change in MKI67 (rho = 0.7, P < 0.001). We identified two samples, which changed from MammaPrint (MP) low-risk to high-risk among 17 pairs with data. One sample's score was on the cutoff for high-risk definition. Interestingly, the second sample also had an increase in Ki67 gene expression and proliferation gene signature score in the post-treatment sample. Overall, ERα/DNA binding interaction regions overlapped significantly more among post-treatment samples as compared to pre-treatment samples (P <0.001). There were 3 samples that increased in MKI67 gene expression after drug exposure. Among these, only the MP low- to high-risk sample had an increase in proliferation gene signature and decrease in ERα/DNA binding interactions. Conclusions: Pre-treatment samples were more variable for both proliferation gene expression signatures and ERα/DNA binding interactions indicating the underlying molecular heterogeneity of the group prior to therapy. This inter-tumor heterogeneity appears to have been lowered by exposure to tamoxifen. Interestingly, not all samples were uniform in their response to tamoxifen exposure as measured by Ki67 and MP scores suggesting samples taken after treatment exposure may be useful for predictive biomarker discovery.

Citation Format: Linn SC, Severson TM, Nevedomskaya E, Peeters J, van Rossum A, Kuilman T, Krijgsman O, Goossens I, Glas A, Koornstra R, Peeper D, Wesseling J, Simon I, Wessels L, Zwart W. Neoadjuvant tamoxifen therapy synchronizes ERα binding and gene expression profiles. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P6-08-06.

Abstract S6-05: High levels of APOBEC3B, a DNA deaminase and an enzymatic source of C-to-T transitions, are a validated marker of poor outcome in estrogen receptor-positive breast cancer

Two recent observations have connected the innate immune DNA cytosine deaminase APOBEC3B to the genetic evolution of breast cancer. First, APOBEC3B was shown to be up-regulated in the majority of breast cancers, and, in breast cancer cell lines, its activity was causally linked to a doubling of the number of C-to-T transitions over time and to a delay in cell cycle progression (1). Second, sequencing of the complete genome of 21 breast cancers independently suggested that APOBEC deaminase activity could be responsible for 2 of 5 mutational imprints identified, which involved clustered (also called kataegis) and dispersed C-to-T transition mutations in the context of 5’TC dinucleotide motifs (2).

In the current study, we addressed a possible association of APOBEC3B expression with outcome in clinical breast cancer. For this we measured using real-time RT-PCR APOBEC3B mRNA levels in 1,491 primary invasive breast cancers and correlated these levels with disease-free survival (DFS), metastasis-free survival (MFS) and overall survival (OS) using univariate and multivariable Cox regression analysis. In addition, we independently validated our findings in available gene expression datasets with appropriate follow-up.

In univariate analyses including all patients, increasing levels of APOBEC3B mRNA analyzed as a continuous variable were significantly associated with shorter DFS, MFS and OS (Hazard Ratio [HR] = 1.29, 1.31 and 1.36, respectively, all P<0.001). To determine the relation of APOBEC3B mRNA expression with the natural course of the disease without the potential confounding effects of systemic adjuvant therapy, we restricted our next analyses to MFS in 829 patients with lymph node-negative disease who had not received any (neo)adjuvant systemic therapy. This analysis showed that APOBEC3B mRNA expression was, in univariate, and in multivariable analysis, including the traditional prognostic factors (age, menopausal status, tumor size, grade and steroid hormone receptors), a marker of pure prognosis specifically in patients with estrogen receptor-positive (ER+) disease (univariate HR = 1.30; P = 0.003; multivariate HR = 1.22, P = 0.042).

To substantiate and validate our findings, we analysed 4 independent available datasets containing in total 5,760 breast cancer cases in which APOBEC3B mRNA expression was measured by probes on microarrays and found that higher APOBEC3B mRNA expression (dichotomised by mean) was significantly associated with poor outcome in all 4 cohorts ([Metabric, 1,491 ER+ cases, HR = 1.82; P<0.001], [Affymetrix compiled dataset-1, 2,407 cases, HR = 2.22; P = 0.001], and [BIG 1-98; 1,207 cases, HR = 2.13; P<0.001 of late recurrence>5 years], and [Affymetrix dataset-2, 643 ER+ cases, HR = 2.04; P = 0.001]).

Altogether, our analyses show that APOBEC3B mRNA - and as a result likely DNA deamination – is a validated predictor of poor outcome in breast cancer, supporting the notion that APOBEC3B is a potentially interesting clinical target for therapeutic intervention to prevent breast cancer progression and metastasis, particularly in ER+ disease.

1. Burns, M.B. et al. Nature 494, 366-70 (2013); 2. Nik-Zainal, S. et al. Cell 149, 979-93 (2012).

Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr S6-05.

Cholinergic chemosensory cells in the auditory tube

The luminal composition of the auditory tube influences its function. The mechanisms involved in the monitoring are currently not known. For the lower respiratory epithelium, such a sentinel role is carried out by cholinergic brush cells. Here, using two different mouse strains expressing eGFP under the control of the promoter of choline acetyltransferase (ChAT), we show the presence of solitary cholinergic villin-positive brush cells also in the mouse auditory tube epithelium. They express the vesicular acetylcholine (ACh) transporter and proteins of the taste transduction pathway such as α-gustducin, phospholipase C beta 2 (PLC(β2)) and transient receptor potential cation channel subfamily M member 5 (TRPM5). Immunoreactivity for TRPM5 and PLCβ2 was found regularly, whereas α-gustducin was absent in approximately 15% of the brush cells. Messenger RNA for the umami taste receptors (TasR), Tas1R1 and 3, and for the bitter receptors, Tas2R105 and Tas2R108, involved in perception of cycloheximide and denatonium were detected in the auditory tube. Using a transgenic mouse that expresses eGFP under the promotor of the nicotinic ACh receptor α3-subunit, we identified cholinoceptive nerve fibers that establish direct contacts to brush cells in the auditory tube. A subpopulation of these fibers displayed also CGRP immunoreactivity. Collectively, we show for the first time the presence of brush cells in the auditory tube. These cells are equipped with all proteins essential for sensing the composition of the luminal microenvironment and for communication of the changes to the CNS via attached sensory nerve fibers.

ER Status and Immune Module Are Central Determinants of HER2 Amplified Infiltrative Breast Carcinoma Prognosis and Pathologic Complete Response

IntroductionGene expression profiling of invasive breast cancer has resulted in highlighting three main categories of breast cancer [luminal, basal, HER2] with very specific features. It has been shown that these subtypes differ in their response to neaoadjuvant systemic treatment, axillary lymph node involvement, metastasis pattern and time to metastasis. The aim of this study was to focus on gene-expression profile analysis of HER2 positive breast carcinoma to assess how the molecular subtype classification applies to the HER2+ve samples; whether subgroups of patients that have different prognosis can be identified and how subgroups can be identified that differ with respect to the likelihood to achieve pathologic complete response (pCR) after neoadjuvant systemic treatment?Materials and MethodWe selected breast carcinomas reported to have a HER2+ve status (IHC and CISH) from patients treated between January 1990 and December 2006 at the Netherlands Cancer Institute. 113 tumor samples were hybridized on the Human Genome Oligo Set Version 3.0 arrays. We identified 205 HER2+ve samples from 4 public microarrays datasets and 33 HER2+ve samples from one public neoadjuvant systemic treatment microarrays dataset.ResultsWe applied the molecular subtype classification to the whole datasets and found major classification instability. 43.3% of the HER2+ve samples were classified as “HER2 subtype”, 26% as “basal-like subtype”, and 21.9% as “luminal-like subtype”. The molecular subtype classification was not correlated to prognosis. Of the prognostic variables tested, only ER status was to the development of distant metastasis (IHC, HR=0.53 [0.36-0.77], p=0.0009). We identified a set of 109 ReporterID's highly enriched in gene ontology annotations link to the Immune Response and correlated to the prognosis of HER2+ve breast carcinoma (Inactivated Immune Module, HR=4.21 [1.94-9.17], logrank pvalue=8.1E-05). The Immune Module Signature was associated with prognosis of the HER2+ve samples in both ER positive and ER negative breast cancer. We validated this Immune Module signature combined with ER status on 205 independent samples.We applied the same classification tree to 33 samples from patients who were assessed for response to neo-adjuvant systemic chemotherapy (Anthracyclin-based) and showed an association between the Immune Module Signature combined with ER status and pCR rate (ER+ve Inactivated Immune Module 12.5% pCR, ER-ve Activated Immune Module 70% pCR, pvalue=0.1).ConclusionWe have shown that HER2+ve breast cancer samples often are not classified as HER2-like by gene expression profiling and that ER (IHC) status determines two major subgroups. We provide new evidence that an immune response relate gene expression classifier has prognostic impact in HER2+ve breast cancer; and that this classifier is potentially correlated to the pCR rate after neoadjuvant systemic treatment in HER2+ve breast cancer.

Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 4037.

Oncogenic pathways impinging on the G2-restriction point

In the absence of mitogenic stimuli, cells normally arrest in G(1/0), because they fail to pass the G1-restriction point. However, abrogation of the G1-restriction point (by loss of the retinoblastoma gene family) reveals a second-restriction point that arrests cells in G2. Serum-starvation-induced G2 arrest is effectuated through inhibitory interactions of p27(KIP1) and p21(CIP1) with cyclins A and B1 and can be reversed through mitogen re-addition. In this study, we have investigated the pathways that allow cell cycle re-entry from this G2 arrest. We provide evidence that recovery from G2 arrest depends on the rat sarcoma viral oncogene (RAS) and phosphatidylinositol-3 kinase pathways and show that oncogenic hits, such as overexpression of c-MYC or mutational activation of RAS can abrogate the G2-restriction point. Together, our results provide new mechanistic insight into multistep carcinogenesis.

[Ocular involvement in systemic Loa-Loa filariasis. Case report and review of the literature]

BACKGROUND

Within the last few years there is more and more evidence for nonspecific ocular symptoms caused by "exotic" pathogens. We herewith report another case of such an infection to underline the increasing importance and diagnostic relevance even of rare diseases.

PATIENT

A 35-year-old female German patient presented with recurrent left retrobulbar "feeling of pressure" after a 6-month-stay in Central Africa 5 years ago. In addition, she reported on repeated swelling of the skin and joints of her hands and arms. Multiple ophthalmologic and rheumatologic investigations had been carried out without diagnostic results. Her ophthalmologist referred her with the presumed diagnosis of a subconjunctival worm.

RESULTS

Slit-lamp biomicroscopy confirmed the original suspicion, and after topical anesthesia a female Loa-Loa worm was easily removed with forceps through a conjunctival incision. The general examination showed symptoms of systemic infection (calabar swelling, eosinophilia). Classification and initiation of treatment with diethylcarbamazine and mebendazole were carried out by the University Institute of Parasitology.

CONCLUSION

Loa-Loa is a parasitic infection endemic in the tropical rain forests of Western, Central, and Eastern Africa. It is transmitted by the Chrysops fly. An increasing number of oculosystemic infections in non-African patients with Loa-Loa are being published. Therefore, any patient with an unclassifiable eye affection should also be investigated for those rare pathogens.

Patients with Multiple Sclerosis and concomitant uveitis/periphlebitis retinae are not distinct from those without intraocular inflammation

Recent models of experimental autoimmune encephalomyelitis (EAE) have indicated that antigens co-expressed in the retina and uvea might be of pathogenetic relevance in Multiple Sclerosis (MS). We investigated the clinical spectrum and magnetic resonance imaging of 11 MS patients with concomitant uveitis, and determined the frequency of clinically silent intraocular inflammation in a prospective series of 50 patients. Two of the 11 patients had panuveitis, seven had anterior, and the remaining two had intermediate uveitis. The onset of uveitis preceded that of neurological symptoms by a mean of 8.5 years (range 1-20). None of the 50 MS patients studied prospectively by using slit lamp examinations and dilated funduscopy showed any evidence of uveitis but six patients had signs of retinal inflammation ("periphlebitis retinae"). Cranial MRI did not reveal "atypical" lesional distribution in MS patients with uveitis or periphlebitis retinae. No correlation between the type of MS and uveitis, or between the degree of neurological disability and the type of uveitis was found.

Experimental strain analysis of Clarens Sandstone colonised by endolithic lichens

Endolithic lichens occur commonly on Clarens Sandstone in South Africa, where they significantly contribute to the weathering of sandstone by means of mechanical and chemical weathering processes. This preliminary investigation reports on the success- ful use of strain gauges in detecting strain differences between sandstone without epilithic lichens and sandstone colonised by the euendolithic lichen Lecidea aff. sarcogynoides Korb. Mechanical weathering, expressed as strain changes, in Clarens Sandstone was studied during the transition from relatively dry winter to wet summer conditions. Daily weathering of sandstone due to thermal expansion and contraction of colonised and uncolonised sandstone could be shown. Our results show that liquid water in sandstone enhances the mechanical weathering of uncolonised Clarens Sandstone while water in the gaseous phase enhances mechanical weathering of sandstone by euendolithic lichens.