Outcome analysis of predictors in locally advanced pancreatic cancer

Guidelines report a wide range of options in locally advanced pancreatic cancer (LAPC): definitive chemotherapy or chemoradiotherapy or the emerging stereotactic body radiotherapy (SBRT) (+/- chemotherapy). On behalf of the AIRO (Italian Association of Radiation Oncology and Clinical Oncology) Gastrointestinal Study Group, we collected retrospective clinical data on 419 LAPC from 15 Italian centers. The study protocol (PAULA-1: Pooled Analysis in Unresectable Locally Advanced pancreatic cancer) was approved by institutional review board of S. Orsola-Malpighi Hospital (201/2015/O/OssN). From this large database we performed tree different studies. The first was a retrospective study about 56 LAPC treated with SBRT at a median biologically equivalent dose of 48 Gy +/- chemotherapy. We demonstrated a statistically significant impact of biologically equivalent dose based on an α/β ratio of 10Gy ≥ 48Gy for local control (LC) (p: 0.045) and overall survival (p: 0.042) in LAPC. The second was a retrospective matched-cohort case-control study comparing SBRT (40 patients) and chemoradiation (40 patients) in LAPC in terms of different endpoints. Our findings suggested an equivalence in terms of most outcomes among the two treatments and an advantage of SBRT in terms of LC (p: 0.017). The third study was a retrospective comparison of definitive chemotherapy, chemoradiotherapy and SBRT (+/- chemotherapy) in terms of different outcomes in LAPC. A predictive model for LC in LAPC was also developed reaching an AUC of 68% (CI 58,7%-77,4%). SBRT treatment emerged as a positive predictive factor for improved LC. Findings deriving from our three studies suggest that SBRT is comparable to standard of care (definitive chemotherapy and chemoradiotherapy) in terms of outcomes. SBRT seems to be an emerging therapeutic option in LAPC significantly improving local control. Furthermore, we have shown the potential of a predictive model for LC. Randomized trials are needed to compare these different therapeutic options in LAPC.

Frequency domain analysis of stationary time series

This thesis provides a necessary and sufficient condition for asymptotic efficiency of a nonparametric estimator of the generalised autocovariance function of a Gaussian stationary random process. The generalised autocovariance function is the inverse Fourier transform of a power transformation of the spectral density, and encompasses the traditional and inverse autocovariance functions. Its nonparametric estimator is based on the inverse discrete Fourier transform of the same power transformation of the pooled periodogram. The general result is then applied to the class of Gaussian stationary ARMA processes and its implications are discussed. We illustrate that for a class of contrast functionals and spectral densities, the minimum contrast estimator of the spectral density satisfies a Yule-Walker system of equations in the generalised autocovariance estimator. Selection of the pooling parameter, which characterizes the nonparametric estimator of the generalised autocovariance, controlling its resolution, is addressed by using a multiplicative periodogram bootstrap to estimate the finite-sample distribution of the estimator. A multivariate extension of recently introduced spectral models for univariate time series is considered, and an algorithm for the coefficients of a power transformation of matrix polynomials is derived, which allows to obtain the Wold coefficients from the matrix coefficients characterizing the generalised matrix cepstral models. This algorithm also allows the definition of the matrix variance profile, providing important quantities for vector time series analysis. A nonparametric estimator based on a transformation of the smoothed periodogram is proposed for estimation of the matrix variance profile.