Aleksander Sadikov

Outcome of small cell lung cancer (SCLC) patients with brain metastases in a routine clinical setting.

Outcome of small cell lung cancer (SCLC) patients with brain metastases in a routine clinical setting Background. Small cell lung cancer (SCLC) represents approximately 13 to 18% of all lung cancers. It is the most aggressive among lung cancers, mostly presented at an advanced stage, with median survival rates of 10 to12 months in patients treated with standard chemotherapy and radiotherapy. In approximately 15-20% of patients brain metastases are present already at the time of primary diagnosis; however, it is unclear how much it influences the outcome of disease according the other metastatic localisation. The objective of this analysis was to evaluate the median survival of SCLC patients treated by specific therapy (chemotherapy and/or radiotherapy) with regard to the presence or absence of brain metastases at the time of diagnosis. Patients and methods. All SCLC patients have been treated in a routine clinical practice and followed up at the University Clinic Golnik in Slovenia. In the retrospective study the medical files from 2002 to 2007 were review. All patients with cytological or histological confirmed disease and eligible for specific oncological treatment were included in the study. They have been treated according to the guidelines valid at the time. Chemotherapy and regular followed-up were carried out at the University Clinic Golnik and radiotherapy at the Institute of Oncology Ljubljana. Results. We found 251 patients eligible for the study. The median age of them was 65 years, majority were male (67%), smokers or ex-smokers (98%), with performance status 0 to 1 (83%). At the time of diagnosis no metastases were found in 64 patients (25.5%) and metastases outside the brain were presented in 153 (61.0%). Brain metastases, confirmed by a CT scan, were present in 34 patients (13.5%), most of them had also metastases at other localisations. All patients received chemotherapy and all patients with confirmed brain metastases received whole brain irradiation (WBRT). The radiotherapy with radical dose at primary tumour was delivered to 27 patients with limited disease and they got 4-6 cycles of chemotherapy. Median overall survival (OS) of 34 patients with brain metastases was 9 months (95% CI 6-12) while OS of 153 patients with metastases in other locations was 11 months (95% CI 10-12); the difference did not reach the level of significance (p = 0.62). As expected, the OS of patients without metastases at the time of primary diagnosis turned out to be significantly better compared to the survival of patients with either brain or other location metastases at the primary diagnosis (15 months vs 9 and 11 months, respectively, p < 0.001). Conclusions. In our investigated population, the prognosis of patients with extensive SCLS with brain metastases at the primary diagnosis treated with chemotherapy and WBRT was not significantly worse compared to the prognosis of patients with extensive SCLC and metastases outside the brain. In extensive SCLC brain metastases were not a negative prognostic factor per se if the patients were able to be treated appropriately. However, the survival rates of extensive SCLC with or without brain metastases remained poor and novel treatment approaches are needed. The major strength of this study is that it has been done on a population of patients treated in a routine clinical setting.

Brain metastases in lung adenocarcinoma: impact of EGFR mutation status on incidence and survival

Abstract Background. The brain represents a frequent progression site in lung adenocarcinoma. This study was designed to analyse the association between the epidermal growth factor receptor (EGFR) mutation status and the frequency of brain metastases (BM) and survival in routine clinical practice. Patients and methods. We retrospectively analysed the medical records of 629 patients with adenocarcinoma in Slovenia who were tested for EGFR mutations in order to analyse the cumulative incidence of BM, the time from the diagnosis to the development of BM (TDBM), the time from BM to death (TTD) and the median survival. Results. Out of 629 patients, 168 (27%) had BM, 90 patients already at the time of diagnosis. Additional 78 patients developed BM after a median interval of 14.3 months; 25.8 months in EGFR positive and 11.8 months in EGFR negative patients, respectively (p = 0.002). EGFR mutations were present in 47 (28%) patients with BM. The curves for cumulative incidence of BM in EGFR positive and negative patients demonstrate a trend for a higher incidence of BM in EGFR mutant patients at diagnosis (19% vs. 13%, p = 0.078), but no difference later during the course of the disease. The patients with BM at diagnosis had a statistically longer TTD (7.3 months) than patients who developed BM later (3.1 months). The TTD in EGFR positive patients with BM at diagnosis was longer than in EGFR negative patients (12.6 vs. 6.8, p = 0.005), while there was no impact of EGFR status on the TTD of patients who developed BM later. Conclusions. Except for a non-significant increase of frequency of BM at diagnosis in EGFR positive patients, EGFR status had no influence upon the cumulative incidence of BM. EGFR positive patients had a longer time to CNS progression. While EGFR positive patients with BM at diagnosis had a longer survival, EGFR status had no influence on TTD in patients who developed BM later during the course of disease.

Bias and pathology in minimax search

This article presents the results of experiments designed to gain insight into the effect of the minimax algorithm on the error of a heuristic evaluation function. Two types of effect of minimax are considered: (a) evaluation accuracy (Are the minimax backed-up values more accurate than the heuristic values themselves?), and (b) decision accuracy (Are moves played by deeper minimax search better than those by shallower search?). The experiments were performed in the King-Rook-King (KRK) chess endgame and in randomly generated game trees. The results show that, counter-intuitively, evaluation accuracy may decline with search depth, whereas at the same time decision accuracy improves with depth. In the article, this is explained by the fact that minimax in combination with a noisy evaluation function introduces a bias into the backed-up evaluations, which masks the evaluation effectiveness of minimax, but this bias still permits decision accuracy to improve with depth. This observed behaviour of minimax in the KRK endgame is discussed in the light of previous studies of pathology in minimax. It is shown that explaining the behaviour of minimax in an actual chess endgame in terms of previously known results requires special care.

Automatic Spiral Analysis for Objective Assessment of Motor Symptoms in Parkinson's Disease

A challenge for the clinical management of advanced Parkinson’s disease (PD) patients is the emergence of fluctuations in motor performance, which represents a significant source of disability during activities of daily living of the patients. There is a lack of objective measurement of treatment effects for in-clinic and at-home use that can provide an overview of the treatment response. The objective of this paper was to develop a method for objective quantification of advanced PD motor symptoms related to off episodes and peak dose dyskinesia, using spiral data gathered by a touch screen telemetry device. More specifically, the aim was to objectively characterize motor symptoms (bradykinesia and dyskinesia), to help in automating the process of visual interpretation of movement anomalies in spirals as rated by movement disorder specialists. Digitized upper limb movement data of 65 advanced PD patients and 10 healthy (HE) subjects were recorded as they performed spiral drawing tasks on a touch screen device in their home environment settings. Several spatiotemporal features were extracted from the time series and used as inputs to machine learning methods. The methods were validated against ratings on animated spirals scored by four movement disorder specialists who visually assessed a set of kinematic features and the motor symptom. The ability of the method to discriminate between PD patients and HE subjects and the test-retest reliability of the computed scores were also evaluated. Computed scores correlated well with mean visual ratings of individual kinematic features. The best performing classifier (Multilayer Perceptron) classified the motor symptom (bradykinesia or dyskinesia) with an accuracy of 84% and area under the receiver operating characteristics curve of 0.86 in relation to visual classifications of the raters. In addition, the method provided high discriminating power when distinguishing between PD patients and HE subjects as well as had good test-retest reliability. This study demonstrated the potential of using digital spiral analysis for objective quantification of PD-specific and/or treatment-induced motor symptoms.

Impact of ERCC1 expression on treatment outcome in small-cell lung cancer patients treated with platinum-based chemotherapy

INTRODUCTION The excision repair cross-complementing 1 (ERCC1) protein is an extensively investigated molecular marker because it may decrease sensitivity to platinum-based chemotherapy. Low ERCC1 expression has already been correlated with better treatment efficacy in non-small-cell lung cancer patients treated with platinum-based chemotherapy. However, the data on a prognostic and/or predictive value of ERCC1 in small-cell lung cancer (SCLC) are still very limited. METHODS This retrospective pilot study evaluated the impact of ERCC1 expression levels on response to first-line platinum-based chemotherapy with or without radiotherapy and survival outcomes of 77 SCLC patients. ERCC1 protein expression was determined immunohistochemically in primary tumour tissue. RESULTS ERCC1 protein expression was positive in 40/77 (51.9%) of our patients. No significant association was found between ERCC1 protein expression and response rate to first-line platinum-based chemotherapy, progression-free survival (PFS), or overall survival (OS), either in the overall population or in patients stratified by disease stage. CONCLUSIONS In our limited group of 77 SCLC patients, ERCC1 protein expression was not found to correlate with either response rate to platinum-based chemotherapy or survival outcomes. Multi-centric prospective trials using a validated method of ERCC1 determination are mandatory in order to obtain a definitive answer on the predictive value of ERCC1 in SCLC.

Learning long-term chess strategies from databases

We propose an approach to the learning of long-term plans for playing chess endgames. We assume that a computer-generated database for an endgame is available, such as the king and rook vs. king, or king and queen vs. king and rook endgame. For each position in the endgame, the database gives the “value” of the position in terms of the minimum number of moves needed by the stronger side to win given that both sides play optimally. We propose a method for automatically dividing the endgame into stages characterised by different objectives of play. For each stage of such a game plan, a stage-specific evaluation function is induced, to be used by minimax search when playing the endgame. We aim at learning playing strategies that give good insight into the principles of playing specific endgames. Games played by these strategies should resemble human expert’s play in achieving goals and subgoals reliably, but not necessarily as quickly as possible.

Barriers and Challenges to Global Clinical Cancer Research

BACKGROUND There are concerns about growing barriers to cancer research. We explored the characteristics of and barriers to global clinical cancer research. METHODS The American Society of Clinical Oncology International Affairs Committee invited 300 selected oncologists with research experience from 25 countries to complete a Web-based survey. Fishers exact test was used to compare answers between participants from high-income countries (HICs) and low- and middle-income countries (LMICs). Barriers to clinical cancer research were ranked from 1 (most important) to 8 (least important). Mann-Whitneys nonparametric test was used to compare the ranks describing the importance of investigated obstacles. RESULTS Eighty oncologists responded, 41 from HICs and 39 from LMICs. Most responders were medical oncologists (62%) at academic hospitals (90%). Researchers from HICs were more involved with academic and industry-driven research than were researchers from LMICs. Significantly higher proportions of those who considered their ability to conduct academic research and industry-driven research over the past 5 years more difficult were from HICs (73% vs. 27% and 70% vs. 30%, respectively). Concerning academic clinical cancer research, a lack of funding was ranked the most important (score: 3.16) barrier, without significant differences observed between HICs and LMICs. Lack of time or competing priorities and procedures from competent authorities were the second most important barriers to conducting academic clinical research in HICs and LMICs, respectively. CONCLUSION Lack of funding, lack of time and competing priorities, and procedures from competent authorities might be the main global barriers to academic clinical cancer research.

Deriving concepts and strategies from chess tablebases

Complete tablebases, indicating best moves for every position, exist for chess endgames. There is no doubt that tablebases contain a wealth of knowledge, however, mining for this knowledge, manually or automatically, proved as extremely difficult. Recently, we developed an approach that combines specialized minimax search with the argument-based machine learning (ABML) paradigm. In this paper, we put this approach to test in an attempt to elicit human-understandable knowledge from tablebases. Specifically, we semi-automatically synthesize knowledge from the KBNK tablebase for teaching the difficult king, bishop, and knight versus the lone king endgame.

Goal-Oriented conceptualization of procedural knowledge

Conceptualizing procedural knowledge is one of the most challenging tasks of building systems for intelligent tutoring. We present an algorithm that enables teachers to accomplish this task semi automatically. We used the algorithm on a difficult king, bishop, and knight versus the lone king (KBNK) chess endgame, and obtained concepts that could serve as textbook instructions. A pilot experiment with students and a separate evaluation of the instructions by experienced chess trainers were deemed very positive.

Learning Positional Features for Annotating Chess Games: A Case Study

By developing an intelligent computer system that will provide commentary of chess moves in a comprehensible, user-friendly, and instructive way, we are trying to use the power demonstrated by the current chess engines for tutoring chess and for annotating chess games. In this paper, we point out certain differences between the computer programs which are specialized for playing chess and our program which is aimed at providing quality commentary. Through a case study, we present an application of argument-based machine learning, which combines the techniques of machine learning and expert knowledge, to the construction of more complex positional features, in order to provide our annotating system with an ability to comment on various positional intricacies of positions in the game of chess.