Outcome Prediction in Mathematical Models of Immune Response to Infection.

Clinicians need to predict patient outcomes with high accuracy as early as possible after disease inception. In this manuscript, we show that patient-to-patient variability sets a fundamental limit on outcome prediction accuracy for a general class of mathematical models for the immune response to infection. However, accuracy can be increased at the expense of delayed prognosis. We investigate several systems of ordinary differential equations (ODEs) that model the host immune response to a pathogen load. Advantages of systems of ODEs for investigating the immune response to infection include the ability to collect data on large numbers of 'virtual patients', each with a given set of model parameters, and obtain many time points during the course of the infection. We implement patient-to-patient variability v in the ODE models by randomly selecting the model parameters from distributions with coefficients of variation v that are centered on physiological values. We use logistic regression with one-versus-all classification to predict the discrete steady-state outcomes of the system. We find that the prediction algorithm achieves near 100% accuracy for v = 0, and the accuracy decreases with increasing v for all ODE models studied. The fact that multiple steady-state outcomes can be obtained for a given initial condition, i.e. the basins of attraction overlap in the space of initial conditions, limits the prediction accuracy for v > 0. Increasing the elapsed time of the variables used to train and test the classifier, increases the prediction accuracy, while adding explicit external noise to the ODE models decreases the prediction accuracy. Our results quantify the competition between early prognosis and high prediction accuracy that is frequently encountered by clinicians.

The Incidence of Cancer Among Acromegaly Patients: Results From the German Acromegaly Registry.

CONTEXT: Acromegaly is a rare disease characterized by high serum levels of GH and IGF-1. Animal studies have demonstrated links between these hormones and cancer, but data regarding cancer incidence among acromegaly patients are inconsistent. Moreover, therapy options have changed considerably since many of the aforementioned data were collected. OBJECTIVE: The objective was to determine whether the overall and site-specific incidence of cancer is comparable to that of the general population. DESIGN AND SETTING: Data from the German Acromegaly Registry for 446 patients (6656 person-years from diagnosis) treated in seven specialized endocrine centers were analyzed. MAIN OUTCOME MEASURE: Standard incidence ratios (SIRs) were calculated as compared to the general population. RESULTS: Overall cancer incidence was slightly but not significantly lower than in the general population (SIR, 0.75; 95% confidence interval, 0.55 to 1.00; P = .051) and was not significantly higher for colorectal, breast, thyroid, prostate, and lung cancers. The SIRs of those with GH in the ranges <1, 1-2.5, and ≥ 2.5 ng/mL were 0.75, 0.44, and 0.92, respectively (P = .94). There was not a significant dependence on normal vs elevated IGF-1 (P = .87), radiation therapy (P = .45), disease duration (P = .96), age at diagnosis (P = .15), or during a period of high GH and IGF-1 from 8 years before to 2 years after diagnosis of acromegaly (P = .41). CONCLUSIONS: Cancer screening strategies need to take incidence into account, which does not seem to be substantially higher in treated acromegaly patients than in the general population for any site of cancer.

Serum levels of acylcarnitines are altered in prediabetic conditions.

OBJECTIVE: The role of mitochondrial function in the complex pathogenesis of type 2 diabetes is not yet completely understood. Therefore, the aim of this study was to investigate serum concentrations of short-, medium- and long-chain acylcarnitines as markers of mitochondrial function in volunteers with normal, impaired or diabetic glucose control. METHODS: Based on a 75 g oral glucose tolerance test, 1019 studied subjects were divided into a group with normal glucose tolerance (NGT; n = 636), isolated impaired fasting glycaemia (IFG; n = 184), impaired glucose tolerance (IGT; n = 87) or type 2 diabetes (T2D; n = 112). Serum concentrations of free carnitine and 24 acylcarnitines were measured by mass spectrometry. RESULTS: Serum levels of acetylcarnitine (C2), propionylcarnitine (C3), octanoylcarnitine (C8), malonylcarnitine/hydroxybutyrylcarnitine (C3DC+C4OH), hexanoylcarnitine (C6), octenoylcarnitine (C8:1), decanoylcarnitine (C10), decenoylcarnitine (C10:1), dodecanoylcarnitine (C12), tetradecenoylcarnitine (C14:1), tetradecadienylcarnitine (C14:2), hydroxytetradecanoylcarnitine (C14OH), hydroxyhexadecanoylcarnitine (C16OH) and octadecenoylcarnitine (C18:1) were significantly different among the groups (all p<0.05 adjusted for age, gender and BMI). Between the prediabetic states C14:1, C14:2 and C18:1 showed significantly higher serum concentrations in persons with IGT (p<0.05). Compared to T2D the IFG and the IGT subjects showed lower serum concentrations of malonylcarnitine/hydroxybutyrylcarnitine (C3DC+C4OH) (p<0.05). CONCLUSION: Alterations in serum concentrations of several acylcarnitines, in particular tetradecenoylcarnitine (C14:1), tetradecadienylcarnitine (C14:2), octadecenoylcarnitine (C18:1) and malonylcarnitine/hydroxybutyrylcarnitine (C3DC+C4OH) are associated not only with T2D but also with prediabetic states.