Predicting antitumor activity of peptides by consensus of regression models trained on a small data sample

Predicting antitumor activity of compounds using regression models trained on a small number of compounds with measured biological activity is an ill-posed inverse problem. Yet, it occurs very often within the academic community. To counteract, up to some extent, overfitting problems caused by a small training data, we propose to use consensus of six regression models for prediction of biological activity of virtual library of compounds. The QSAR descriptors of 22 compounds related to the opioid growth factor (OGF, Tyr-Gly-Gly-Phe-Met) with known antitumor activity were used to train regression models: the feed-forward artificial neural network, the k-nearest neighbor, sparseness constrained linear regression, the linear and nonlinear (with polynomial and Gaussian kernel) support vector machine. Regression models were applied on a virtual library of 429 compounds that resulted in six lists with candidate compounds ranked by predicted antitumor activity. The highly ranked candidate compounds were synthesized, characterized and tested for an antiproliferative activity. Some of prepared peptides showed more pronounced activity compared with the native OGF; however, they were less active than highly ranked compounds selected previously by the radial basis function support vector machine (RBF SVM) regression model. The ill-posedness of the related inverse problem causes unstable behavior of trained regression models on test data. These results point to high complexity of prediction based on the regression models trained on a small data sample.

[QT interval dispersion in hypertensive diabetics and in patients with hypertension with chronic heart failure without diabetes]

AIM

Our aim was to: 1. compare QT dispersion from routine ECG in diabetic and no-diabetic patients with congestive heart failure, 2. describe associations between QT dispersion and circadian blood (BP) pressure variation in type 2 diabetic patients with congestive heart failure (CHF).

PATIENTS AND METHODS

122 patients admitted to hospital due to CHF in the period between years 2000-2001 have been divided into 2 groups: group 1:70 patients (m: 40, f: 30, mean age 64.7 +/- 9 years) with type II diabetes mellitus (DM), group 2:52 patients (m: 28, f:24, mean age 62.5 +/- 10.9 years) without DM. Diagnosis of CHF was made clinically and proved by ECG and ECHO (EF < 40%), DM was defined clinically or by using oral glucose tolerance test (75 g glucose, 2 h blood glucose > 11.1 mmol/l). The QT interval was measured from the beginning of the QRS complex to the end of the T wave from routine 12-lead ECG. QT intervals were corrected for heart rate using Bazett's formula. QT dispersion (QTd) and rate corrected QT dispersion (QTc) were defined as the difference between the maximum and minimum QT and QTc intervals, respectively. Ambulatory blood pressure (AMBP) was measured by an oscillometic technique. Diabetic patients with CHF were divided both according to below and above the median QTc dispersion (65 ms).

STATISTICAL ANALYSIS

Chi-square and Student's t-test. Significant differences were assumed of p < 0.05.

RESULTS

Both groups were matched by gender, age, duration and intensity of hypertension, the presence and intensity of obesity, hyperlipidemia (TC, TG, LDL-C, HDL-C) and smoking habits. Diabetic patients with CHF had significantly longer QTc interval (maximum and minimum), QT dispersion and QTc dispersion compared with non-diabetic patients with CHF. Diabetic patients with CHF with QTc dispersion > 65 ms had significantly higher night systolic (133 +/- 14 vs. 112 +/- 14) and diastolic (80 +/- 11 vs. 65 +/- 6) BP and significantly higher night/day ratio for both systolic (0.94 +/- 0.05 vs. 0.86 +/- 0.06) and diastolic (0.89 +/- 0.07 vs. 0.80 +/- 0.05) compared with diabetic patients with CHF with QTc dispersion < 65 ms.

CONCLUSION

Diabetic patients with CHF are higher risk than non-diabetic. Our data describe both factors related to cardiovascular risk in diabetic patients with CHF-prolongation of the QT and QTc dispersion and reduced nocturnal blood pressure.

[Effect of chronic Chlamydia infection with non-specific inflammation on cardiovascular complications in acute myocardial infarct]

It is known that local and systemic inflammatory processes play an important role in the genesis and development of atheroclerotic lesions and in the pathophysiology of acute coronary syndromes. This hypothesis is supported by findings of elevated parameters of the "inflammatory" reaction in the affected blood vessels but also in the blood of atherosclerotic patients. Known risk factors do not explain quite satisfactorily epidemiological cardiovascular phenomena and different manifestations of coronary heart disease. It is very probable that also Chlamydia pneumoniae is a risk factor. This assumption is based on evaluation of seroepidemiological data, examination of atherosclerotic plaques not only in humans but also in animal models with chlamydial infection. Based on retrospective and prospective evaluation of case-records the authors analyzed the incidence of cardiovascular complications in 83 patients with acute myocardial infarction (AIM), incl. 51 patients (31 men and 20 women, mean age 64.4 +/- 3.4 years who had a non-specific inflammation and chlamydial infection, and 32 patients (24 men and 8 women, mean age 64.7 +/- 3.6 years) who had chlamydial infections but no non-specific inflammation (in the blood). These patients were selected from all patients hospitalized during 1998-2001. When diagnosing acute myocardial infarction we applied WHO criteria, and the presence of at least two of three criteria was necessary: a history of prolonged (more than 20 min). stenocardia, electrocardiographic changes typical for ischaemia and/or necrosis and elevation of myocardial enzymes in serum, Non-specific inflammatory activity was present in patients (i.e. positive) if the following laboratory parameters were recorded: C-reactive protein > 5 mg/l assessed by the radial immunodiffusion method; fibrinogen > 4 mg/l assessed by the coagulation method according to Claus; leukocytes > 9.6 x 10(3)/microliter, leukocytes were counted automatically in a Coulter chamber; lymphocytes > 3.4 x 10(3)/microliter. Red cell sedimentation rate > 20 mm/hour. The activity was evaluated as positive when all parameters were elevated. The presence of chronic infection with Chlamydia pneumoniae was assessed qualitatively by antibody positivity (IgG) in serum using the microimmunoflurescent method (using a set from Labsystems Co.). The incidence of associated risk factors (obesity, smoking, diabetes, hyperlipidaemia and hypertension) is higher in the sub-group of patients with Chlamydia infections without inflammation, however, the difference is not statistically significant. The incidence of cardiovascular attacks was higher in the sub-group of patients with chlamydial infection and concurrent inflammation as compared with the sub-group of patients with chlamydial infection without inflammation. In case of re-infarction of the myocardium, a sudden cerebrovascular attack, death and arrhythmia the difference was statistically significant, while in case of cardiac failure and cardiogenic shock the difference was not significant. Patients with acute myocardial infarction with chlamydial infection and a concurrent non-specific inflammation had to be treated more often by combined (i.e. more intense) treatment, thrombolytic treatment, PTCA and surgery (bypass) of the coronary vessels as compared with patients with Chlamydia infections but without inflammation. The authors assume therefore that not only different risk factors but also the effect of non-specific inflammation and Chlamydia infection contribute towards the increased number of cardiovascular postinfarction complications. Therefore a therapeutic approach involving eradication of infection and suppression of the inflammatory reaction should be considered.

Is left ventricular hypertrophy a risk factor in hypertensive patients?

BACKGROUND

Left ventricular hypertrophy (LVH) is supposed to be a risk factor of cardiovascular (CV) complications in hypertensive patients.

AIM

To compare clinical events in hypertensives with and without LVH.

PATIENTS AND METHODS

319 hypertensives with LVH (mean age 64.1+/-10.6 ys) and 177 hypertensives without LVH (mean age 62.5+/-11.3 ys). LVH defined by echo Penn convention as left ventricular mass index >134 g/m2 in men and >110 g/m2 in women. Clinical events--heart failure (EF<40 %), left ventricular diastolic dysfunction (echo-doppler: transmitral-flow where peak A>peak E), myocardial infarction (history, ECG, cardiac enzymes), chronic atrial fibrillation (more than 2 weeks duration), mitral regurgitation (echo) and renal involvement (creatininemia>120 micromol/l). The two groups of hypertensives were matched by demographic criteria, duration and intensity of hypertension, obesity, diabetes mellitus, lipid serum levels and smoking habits.

RESULTS

There were statistically significant at least p<0.05 more CV events (heart failure, left ventricular diastolic dysfunction, myocardial infarction, chronic atrial fibrillation, and mitral regurgitation cases) and renal involvement in LVH-positive patients than in LVH-negative patients.

CONCLUSION

LVH is a strong risk factor for clinical events in hypertensives, which necessitates their more intensive treatment, mainly with drugs producing also LVH regression. (Tab. 5, Ref. 48.)

[The role of infection and inflammation in the pathogenesis of atherosclerosis]

Pathogenesis of the atherosclerotic process is deemed as multifactorial. To the most important risk factors, besides certain family predisposition, there belongs hypercholesterolemia, arterial hypertension, obesity, diabetes mellitus, smoking and others. In the last years there are more and more data about the role of inflammation and infection in the whole development of atherosclerosis. The witness for this hypothesis is the findings of high parameters of inflammation in involved vessels as well as in the blood of atherosclerosis suffering persons. Opinions about the inflammation theory appear from the 90th. Local sterile inflammation in the subendotelium of the middle and big arteries has been proved to consist of specific immune reaction (activation of the T-lymphocytes) as well as nonspecific characteristic by elevated monocytes in the artery wall during the whole process of atherogenesis. Inflammation in the plaque can trigger and hold several factors engaged in the atherosclerotic process, such as oxidized LDL cholesterol, elevated production of various superoxides, activated macrophages, activated T-lymphocytes, cytokines (IL-1, IL-6, interferon gamma) and lipoprotein Lp (a). In this inflammation process levels of CRP (acute phase protein), fibrinogen and erythrocyte sedimentation are elevated as a reaction of the organism to nonspecific chronic infections. Because of this it is thought that elevated fibrinogen and erythrocyte sedimentation are markers of the cardiovascular risk. Some papers deal with antiinflammatory effects of statins, because these lower CRP levels so they also lower atherosclerotic risk through not only lowering of cholesterol levels. Also asprine, as an antiinflammation agent, changing the CRP levels, would be of benefit for patients with vascular disease because its antiaggregation and antiinflammatory effects. ACE inhibitors are also antiinflamatory through blocking of tissue production of angiotensin II (artery wall and atherosclerotic plaque). Enzymatic inhibitors changing angiotensin can also have a partial antiinflammatory effect. The infection theory is supported also by tracing of some microorganisms in the atherosclerotic plaque or in the blood, as e.g. Helicobacter pylori or Chlamydia pneumoniae; to the autoimmune origin is indicated the presence of the specific immunity reaction against heat shock proteins (HSP) or oxidized LDL. This infection theory offers new therapy possibilities. Therefore eradication for example by antibiotics can lead to stabilization of the atherosclerotic plaque with positive consequences, as it was discovered by many studies.

Relation of left ventricular hypertrophy to cardiovascular complications in diabetic hypertensives

The presence of diabetes mellitus and other risk factors of atherosclerosis, such as obesity, smoking and hyperlipidemia, in hypertensive patients makes the prognosis worse. Authors compared the clinical findings in diabetic hypertensive patients with and without left ventricular hypertrophy, the presence of which was diagnosed and defined by echocardiography. The study is based on the analysis of hospital records of 115 hypertensive patients treated at our department during the period 1998-1999. Left ventricular hypertrophy (LVH) was defined by echocardiography as left ventricular mass index > 134 g/m2 in men and > 110 g/m2 in women. Left ventricular hypertrophy was found in 79 patients (mean age 64.6 ys) but not in 36 patients (mean age 63.3 ys). Both groups were matched as to age and sex, intensity and duration of hypertension and diabetes, obesity, smoking and hyperlipidemia. In LVH-positive patients, there was a statistically significant incidence of heart failure, mitral regurgitation and renal involvement and a more non-significant incidence of left ventricular diastolic dysfunction, myocardial infarction, chronic atrial fibrillation and stroke than in LVH-negative ones. Left ventricular hypertrophy usually complicates the course of hypertension. Authors recommend to investigate the presence of left ventricular hypertrophy in hypertensives as it carries a much more complicated course of the disease. (Tab. 5, Ref. 28.)

[Left ventricular hypertrophy in hypertension]

Left ventricular hypertrophy LVH is supposed to be a useful marker of cardiovascular complications during the course of hypertension. Occurrence of other risk factors of atherosclerosis in these hypertensive patients such as hyperlipidemia and smoking deteriorate the prognosis too. The authors compared clinical findings in hypertensive patients with and without left ventricular hypertrophy defined by echocardiography. Hospital records of 185 hypertensive patients treated at our medical department during years 1996-1999 were analysed. Left ventricular hypertrophy was defined by echocardiography (Penn convention) as left ventricular mass index > 134 g/m2 in men and > 110 g/m2 in women. Presence of LVH was found in 109 patients (mean age 66.7 years), absence of LVH in 76 patients (mean age 64.7 years). Both groups of hypertensive patients were matched by demographic parameters by the presence of hyperlipidemia and by smoking habits. Hypertensive patients with diabetes mellitus and obesity were excluded. They were statistically significant in the incidence of heart failure, myocardial infarction, renal failure and mitral regurgitation, and non-significant in the incidence of left ventricular diastolic dysfunction. There were more cardiovascular complications in LVH-positive patients than in those with LVH-negative findings. The incidence of stroke was slightly higher in LVH-negative patients. Left ventricular hypertrophy in patients with hypertension brings usually a complicated course of the disease. The authors recommend to examine the patients with arterial hypertension for the presence of left ventricular hypertrophy as it complicates the course of the disease significantly. (Tab. 3, Fig. 2, Ref. 26.)

[QT dispersion intervals in hypertensives with left ventricular hypertrophy]

Left ventricular hypertrophy is an important risk factor of cardiovascular complications during the course of hypertension. Increased QT dispersion is associated with sudden cardiac death in congestive heart failure and in other cardiovascular diseases. Our aim was to compare QT dispersion from routine ECG in hypertensive patients with and without left ventricular hypertrophy defined by echocardiography. Authors examined 71 hypertensives treated in our medical department. Left ventricular hypertrophy was defined by echocardiography (Penn convention) as left ventricular mass index > 134 g/m2 in men and > 110 g/m2 in women. QT dispersion was defined from routine ECG (QTmax - QTmin). Presence of LVH was found in 26 patients (mean age 59.3 years), absence of LVH in 45 patients (mean age 57.8 years). Hypertensives with secondary hypertension, hypertrophic cardiomyopathy, sings of ischemia in ECG, arrhythmias, myocardial infarction, heart failure, diabetes mellitus and patients treated by antiarrhythmic drugs of the Ic and III groups were excluded. Both groups of hypertensives were matched by demographic parameters, and by the presence of hypertension, obesity, hyperlipidemia and smoking habites. There were statistically significant longer QT dispersion and QTc dispersion (59.0 +/- 20.1 ms, 64.0 +/- 23.7 ms) in LVH-positive patients than in LVH-negative once (43.2 +/- 9.5 ms, 48.4 +/- 11.1 ms). Left ventricular hypertrophy in patients with hypertension brings usually a complicated course of the disease. Authors recommend to look after left ventricular hypertrophy presence in hypertensives as it carries much more complicated course of the disease. Measurment of QT dispersion adds farther stratificational information to these patients.