Bayesian and survival model for tumor relapse status and disease-specific survival, with applications to breast cancer

Breast cancer is the most common non-skin cancer and the second leading cause of cancer-related death in women in the United States. Studies on ipsilateral breast tumor relapse (IBTR) status and disease-specific survival will help guide clinic treatment and predict patient prognosis.^ After breast conservation therapy, patients with breast cancer may experience breast tumor relapse. This relapse is classified into two distinct types: true local recurrence (TR) and new ipsilateral primary tumor (NP). However, the methods used to classify the relapse types are imperfect and are prone to misclassification. In addition, some observed survival data (e.g., time to relapse and time from relapse to death)are strongly correlated with relapse types. The first part of this dissertation presents a Bayesian approach to (1) modeling the potentially misclassified relapse status and the correlated survival information, (2) estimating the sensitivity and specificity of the diagnostic methods, and (3) quantify the covariate effects on event probabilities. A shared frailty was used to account for the within-subject correlation between survival times. The inference was conducted using a Bayesian framework via Markov Chain Monte Carlo simulation implemented in softwareWinBUGS. Simulation was used to validate the Bayesian method and assess its frequentist properties. The new model has two important innovations: (1) it utilizes the additional survival times correlated with the relapse status to improve the parameter estimation, and (2) it provides tools to address the correlation between the two diagnostic methods conditional to the true relapse types.^ Prediction of patients at highest risk for IBTR after local excision of ductal carcinoma in situ (DCIS) remains a clinical concern. The goals of the second part of this dissertation were to evaluate a published nomogram from Memorial Sloan-Kettering Cancer Center, to determine the risk of IBTR in patients with DCIS treated with local excision, and to determine whether there is a subset of patients at low risk of IBTR. Patients who had undergone local excision from 1990 through 2007 at MD Anderson Cancer Center with a final diagnosis of DCIS (n=794) were included in this part. Clinicopathologic factors and the performance of the Memorial Sloan-Kettering Cancer Center nomogram for prediction of IBTR were assessed for 734 patients with complete data. Nomogram for prediction of 5- and 10-year IBTR probabilities were found to demonstrate imperfect calibration and discrimination, with an area under the receiver operating characteristic curve of .63 and a concordance index of .63. In conclusion, predictive models for IBTR in DCIS patients treated with local excision are imperfect. Our current ability to accurately predict recurrence based on clinical parameters is limited.^ The American Joint Committee on Cancer (AJCC) staging of breast cancer is widely used to determine prognosis, yet survival within each AJCC stage shows wide variation and remains unpredictable. For the third part of this dissertation, biologic markers were hypothesized to be responsible for some of this variation, and the addition of biologic markers to current AJCC staging were examined for possibly provide improved prognostication. The initial cohort included patients treated with surgery as first intervention at MDACC from 1997 to 2006. Cox proportional hazards models were used to create prognostic scoring systems. AJCC pathologic staging parameters and biologic tumor markers were investigated to devise the scoring systems. Surveillance Epidemiology and End Results (SEER) data was used as the external cohort to validate the scoring systems. Binary indicators for pathologic stage (PS), estrogen receptor status (E), and tumor grade (G) were summed to create PS+EG scoring systems devised to predict 5-year patient outcomes. These scoring systems facilitated separation of the study population into more refined subgroups than the current AJCC staging system. The ability of the PS+EG score to stratify outcomes was confirmed in both internal and external validation cohorts. The current study proposes and validates a new staging system by incorporating tumor grade and ER status into current AJCC staging. We recommend that biologic markers be incorporating into revised versions of the AJCC staging system for patients receiving surgery as the first intervention.^ Chapter 1 focuses on developing a Bayesian method to solve misclassified relapse status and application to breast cancer data. Chapter 2 focuses on evaluation of a breast cancer nomogram for predicting risk of IBTR in patients with DCIS after local excision gives the statement of the problem in the clinical research. Chapter 3 focuses on validation of a novel staging system for disease-specific survival in patients with breast cancer treated with surgery as the first intervention. ^

Design, Calibration, and Evaluation of Depth-of-Interaction-Capable PET Detector Modules

High-resolution, small-bore PET systems suffer from a tradeoff between system sensitivity, and image quality degradation. In these systems long crystals allow mispositioning of the line of response due to parallax error and this mispositioning causes resolution blurring, but long crystals are necessary for high system sensitivity. One means to allow long crystals without introducing parallax errors is to determine the depth of interaction (DOI) of the gamma ray interaction within the detector module. While DOI has been investigated previously, newly available solid state photomultipliers (SSPMs) well-suited to PET applications and allow new modules for investigation. Depth of interaction in full modules is a relatively new field, and so even if high performance DOI capable modules were available, the appropriate means to characterize and calibrate the modules are not. This work presents an investigation of DOI capable arrays and techniques for characterizing and calibrating those modules. The methods introduced here accurately and reliably characterize and calibrate energy, timing, and event interaction positioning. Additionally presented is a characterization of the spatial resolution of DOI capable modules and a measurement of DOI effects for different angles between detector modules. These arrays have been built into a prototype PET system that delivers better than 2.0 mm resolution with a single-sided-stopping-power in excess of 95% for 511 keV g's. The noise properties of SSPMs scale with the active area of the detector face, and so the best signal-to-noise ratio is possible with parallel readout of each SSPM photodetector pixel rather than multiplexing signals together. This work additionally investigates several algorithms for improving timing performance using timing information from multiple SSPM pixels when light is distributed among several photodetectors.

Calibration of tracer concentration measurement and results

A reliable assessment of relevant substance flows is very important for environmental risk assessments and efficiency analysis of measures to reduce or avoid emissions of micropollutants like drugs to water systems. Accordingly, a detailed preparation of monitoring campaigns should include an accuracy check for the sampling configuration to prove the reliability of the monitoring results and the subsequent data processing. The accuracy of substance flow analyses is expected to be particularly weak for substances having high short-term variations of concentrations in sewage. This is especially the case linked to the observation of substance flows close to source in waste water systems. The verification of a monitoring configuration in a hospital sewer in Luxembourg is in the centre of interest of the case study presented here. A tracer test in the sewer system under observation is an essential element of the suggested accuracy check and provides valuable information for an uncertainty analysis. The results illustrate the importance of accuracy checks as an essential element of the preparation of monitoring campaigns. Moreover the study shows that continuous flow proportional sampling enables a representative observation of short-term peak loads of the iodinated x-ray contrast media iobitridol close to source.

Maximum and mean diameter record of Orbulina universa from DSDP Hole 47-397 of Cape Bojador, eastern North Atlantic

Measurements of the diameter of O. universa carried out on 30 specimens from 39 samples covering a sediment thickness of 78 m and going back in time to approximately 750 000 y resulted in the construction of a curve of the mean diameter and a curve of the maximum diameter. Both curves, as well as those calculated with the running-averages technique, display cyclic fluctuations with durations of the order of 100 000 y and downwards decreasing amplitudes. These curves are compared with a carbonate curve (on bulk sediment) and an isotopic curve (on benthic foraminifers) obtained from the same set of samples. Correlations are fair to good, but a timelag is noticed between the isotopic curve and the faunal (O. universa mean diameter) curve, with the isotopic signal coming first, in the middle part of the Brunhes Epoch. Biostratigraphic calibration to the paleomagnetic record is provided by four datum planes (two based on calcareous nannofossils, two on diatoms) identified in the succession. Changes recorded in test porosity seem to be less meaningful than changes in test size.

Calibration and application of B/Ca, Cd/Ca, and d11B in Neogloboquadrina pachyderma (sinistral)

The North Atlantic and Norwegian Sea are prominent sinks of atmospheric CO2 today, but their roles in the past remain poorly constrained. In this study, we attempt to use B/Ca and d11B ratios in the planktonic foraminifera Neogloboquadrina pachyderma (sinistral variety) to reconstruct subsurface water pH and pCO2 changes in the polar North Atlantic during the last deglaciation. Comparison of core-top results with nearby hydrographic data shows that B/Ca in N. pachyderma (s) is mainly controlled by seawater [B(OH)4]?/[HCO3]? with a roughly constant partition coefficient (KD =([B/Ca]of CaCO3)/([B(OH)4]-/[HCO3]-)of seawater) of 1.48 ± 0.15 * 10**-3 (2sigma), and d11B in this species is offset below d11B of the borate in seawater by 3.38 ± 0.71 per mil (2sigma). These values represent our best estimates with the sparse available hydrographic data close to our core-tops. More culturing and sediment trap work is needed to improve our understanding of boron incorporation into N. pachyderma (s). Application of a constant KD of 1.48 * 10**-3 to high resolution N. pachyderma (s) B/Ca records from two adjacent cores off Iceland shows that subsurface pCO2 at the habitat depth of N. pachyderma (s) (~50 m) generally followed the atmospheric CO2 trend but with negative offsets of ~10-50 ppmv during 19-10 ka. These B/Ca-based reconstructions are supported by independent estimates from low-resolution d11B measurements in the same cores. We also calibrate and apply Cd/Ca in N. pachyderma (s) to reconstruct nutrient levels for the same down cores. Like today's North Atlantic, past subsurface pCO2 variability off Iceland was significantly correlated with nutrient changes that might be linked to surface nutrient utilization and mixing within the upper water column. Because surface pCO2 (at 0 m water depth) is always lower than at deeper depths and if the application of a constant KD is valid, our results suggest that the polar North Atlantic has remained a CO2 sink during the calcification seasons of N. pachyderma (s) over the last deglaciation.

Radiolarian stratigraphy across the Oligocene/Miocene transition in low latitudes

Recently the International Union of Geological Sciences (Commission on Stratigraphy, Working Group on the Paleogene/Neogene Boundary) proposed that the Oligocene/Miocene boundary be placed at the base of Chron C6Cn2n at 23.8 Ma on the Cande and Kent (1992) magnetic time scale, where it is approximated by planktic foraminifera at the first occurrence of Globorotulia kugleri, and by calcareous nannofossils at the last occurrence of Sphenolithus ciperoensis and the first and last occurrences of Sphenolithus delphix and S. capricornutus. Herein we show that, in terms of radiolarians, the base of Chron C6Cn2n can be correlated with the upper part of the Lychnocanoma elongata Zone between the last occurrence of Artophormis gracilis (23.94 Ma) and the first occurrence of Cyrtocapsella tetrapera (23.69 Ma). Since the proposed stratotype at Lemme-Carrosio (Italy) does not contain radiolarians at the boundary, we re-examined 13 DSDP sites and established the stratigraphic sequence of 29 first and last radiolarian occurrences and one evolutionary transition across the boundary. Nine of these sites contain both calcareous and siliceous microfossils and thus allow for an integrated biostratigraphy. Paleomagnetic stratigraphy is not available for any of the DSDP cores examined. However, use of Hodell and Woodruff's (1994) strontium isotope curve from DSDP Site 289 has permitted calibration of several low latitude microfossil datum levels against the geomagnetic polarity scale. Two new species, Lychnocanoma apodora and Eucyrtidium plesiodiaphanes, are described.

Miocene strontium isotope record of DSDP Hole 30-289

A composite strontium isotopic seawater curve was constructed for the Miocene between 24 and 6 Ma by combining 87Sr/86Sr measurements of planktonic foraminifera from Deep Sea Drilling Project sites 289 and 588. Site 289, with its virtually continuous sedimentary record and high sedimentation rates (26 m/m.y.), was used for constructing the Oligocene to mid-Miocene part of the record, which included the calibration of 63 biostratigraphic datums to the Sr seawater curve using the timescale of Cande and Kent (1992 doi:10.1029/92JB01202). Across the Oligocene/Miocene boundary, a brief plateau occurred in the Sr seawater curve (87Sr/86Sr values averaged 0.70824) which is coincident with a carbon isotopic maximum (CM-O/M) from 24.3 to 22.6 Ma. During the early Miocene, the strontium isotopic curve was marked by a steep rise in 87Sr/86Sr that included a break in slope near 19 Ma. The rate of growth was about 60 ppm/m.y. between 22.5 and 19.0 Ma and increased to over 80 ppm/m.y. between 19.0 and 16 Ma. Beginning at ~16 Ma (between carbon isotopic maxima CM3 and CM4 of Woodruff and Savin (1991 doi:10.1029/91PA02561)), the rate of 87Sr/86Sr growth slowed and 87Sr/86Sr values were near constant from 15 to 13 Ma. After 13 Ma, growth in 87Sr/86Sr resumed and continued until ~9 Ma, when the rate of 87Sr/86Sr growth decreased to zero once again. The entire Miocene seawater curve can be described by a high-order function, and the first derivative (d87Sr/86Sr/dt) of this function reveals two periods of increased slope. The greatest rate of 87Sr/86Sr change occurred during the early Miocene between ~20 and 16 Ma, and a smaller, but distinct, period of increased slope also occurred during the late Miocene between ~12 and 9 Ma. These periods of steepened slope coincide with major phases of uplift and denudation of the Himalayan-Tibetan Plateau region, supporting previous interpretations that the primary control on seawater 87Sr/86Sr during the Miocene was related to the collision of India and Asia. The rapid increase in 87Sr/86Sr values during the early Miocene from 20 to 16 Ma imply high rates of chemical weathering and dissolved riverine fluxes to the oceans. In the absence of another source of CO2, these high rates of chemical weathering should have quickly resulted in a drawdown of atmospheric CO2 and climatic cooling through a reversed greenhouse effect. The paleoclimatic record, however, indicates a warming trend during the early Miocene, culminating in a climatic optimum between 17 and 14.5 Ma. We suggest that the high rates of chemical erosion and warm temperatures during the climatic optimum were caused by an increase in the contribution of volcanic CO2 from the eruption of the Columbia River Flood Basalts (CRFB) between 17 and 15 Ma. The decrease in the rate of CRFB eruptions at 15 Ma and the removal of atmospheric carbon dioxide by increased organic carbon burial in Monterey deposits eventually led to cooling and increased glaciation between ~14.5 and 13 Ma. The CRFB hypothesis helps to explain the significant time lag between the onset of increased rates of organic carbon burial in the Monterey at 17.5 Ma (as marked by increased delta13C values) and the climatic cooling and glaciation during the middle Miocene (as marked by the increase in delta18O values), which did not begin until ~14.5 Ma.

Contiunous surface water uncorrected and quench-corrected active fluorescence and derived chlorophyll a concentration using calibration by HPLC pigments during POLARSTERN cruise ANT-XXV/1

The relationship between phytoplankton assemblages and the associated optical properties of the water body is important for the further development of algorithms for large-scale remote sensing of phytoplankton biomass and the identification of phytoplankton functional types (PFTs), which are often representative for different biogeochemical export scenarios. Optical in-situ measurements aid in the identification of phytoplankton groups with differing pigment compositions and are widely used to validate remote sensing data. In this study we present results from an interdisciplinary cruise aboard the RV Polarstern along a north-to-south transect in the eastern Atlantic Ocean in November 2008. Phytoplankton community composition was identified using a broad set of in-situ measurements. Water samples from the surface and the depth of maximum chlorophyll concentration were analyzed by high performance liquid chromatography (HPLC), flow cytometry, spectrophotometry and microscopy. Simultaneously, the above- and underwater light field was measured by a set of high spectral resolution (hyperspectral) radiometers. An unsupervised cluster algorithm applied to the measured parameters allowed us to define bio-optical provinces, which we compared to ecological provinces proposed elsewhere in the literature. As could be expected, picophytoplankton was responsible for most of the variability of PFTs in the eastern Atlantic Ocean. Our bio-optical clusters agreed well with established provinces and thus can be used to classify areas of similar biogeography. This method has the potential to become an automated approach where satellite data could be used to identify shifting boundaries of established ecological provinces or to track exceptions from the rule to improve our understanding of the biogeochemical cycles in the ocean.

Table 1 : AMS-14C data used for the reconstruction of the Wismar Bay shoreline displacement curve

The Baltic coast of Mecklenburg-Vorpommern is located in the transition Zone between the region of Fennoscandian Uplift and the Central European Depression. In relation to the eustatic sea-level rise, the northeast coast shows a slower inundation, while for the southwestern area a faster transgression is indicated, which can be attributed to crustal movements. To determine the spatial and temporal differences since the onset of the Littorina Transgression, three relative sea-level curves have been established along a transect parallel to the gradient of upliftlsubsidence. The Wismar Bay area is one endpoint of the transect demonstrating today 10 Abb., 2 Tab. a relative sea-level rise of 1.4 mm/a. To determine the relative sea-level curve for the Wismar Bay, two sites were investigated on Rustwerder Spit (Poel) and Redentin. They provided reliable depth-age data, while the stratigraphy was additionally supported by lithological/geochemical, pollen, diatom and macrofossil data. Additional evidence was provided by archaeological submarine surveys and excavations. Comparing the new relative sea-level curve with a curve from the Vorpommern coast, it can be shown that for the period from 4000 cal BC until present, the differences between the two curves are caused by a constant neotectonic movement, while for the older periods an increasing isostatic component must be taken into account.

Tab. 1: Abflußwerte der Pegelstation Vernagtbach

This paper describes the extraordinary difficulties met in gauging glacier streams and establishes criteria that have to be fulfilled for a correct determination of glacier run off. Discussion of the hydraulic principles of gauging in tranquil and rapid flow shows why the latter is to be favored for glacier streams.The concept of the station at Vernagtbach, the original stream bed, considerations ofproper dimensioning and characteristics of the construction of the gauging channel aredescribed in detail. A brief history is given of planning, organization and technology involved.The water level is recorded by two floats and one pneumatie gauge which together provide uninterrupted and trouble-free records. The rating curve was determined with current meters and shows an unambiguous relation between water level and discharge with little scattering of calibration values. Temporary changes of flow conditions upstream of the station do not have any noticeable effect on the water level in the channel.The discharge records of the summers 1974-1976 are presented and discussed. The maximum mean summer discharge was 1.33 m**3/s in June-September 1975; extreme mean monthly discharge was found in July 1976 with 2.51 m**3/s, the highest daily mean was 4.76 m**3/s and the maximum hourly mean was found at 7.23 m**3/s. The discharge conditions of the summers of 1974 and 1975 are very similar, while in the summer of 1976 they differed completely as far as seasonal and mean daily hydrographs are concerned: in 1975, 42 % of summer discharge was recorded in June and July compared to 76 % in 1976. The analysis of the hydrographs gives valuable clues to the mass and heat balances of the glacier.

Strontium isotope ratios of middle Eocene to Oligocene sediments from Maud Rise, Antarctica

A 87Sr/86Sr isotope curve of the middle Eocene to Oligocene was produced from analysis of foraminifera in Ocean Drilling Program Hole 689B, Maud Rise, near the coast of Antarctica. Sediments from the hole are well preserved with no evidence of diagenetic alteration. The sequence is nearly complete from 46.3 to 24.8 Ma, with an average sampling interval of 166 kyr. Excellent magnetostratigraphy in Hole 689B allows calibration to the geomagnetic polarity time scale of Cande and Kent (1992). Marine strontium isotopic ratios were nearly stable from 46.3 to 35.5 Ma, averaging near 0.70773, after which they began to increase. A slow increase began after 40.4 Ma, rising at a rate of only about 8*10**-6/m.y. from base values of 0.707707. From 35.5 Ma to 24.8 Ma the average slope increased to 40*10**-6/m.y. The slope remained constant at least until 24.8 Ma, when the record becomes discontinuous owing to unconformities. We evaluate several possible controls on the marine strontium isotope curve that could have led to the observed growth in 87Sr/86Sr ratios near the Eocene/Oligocene boundary. Three mechanisms are considered, including the onset of Antarctic glaciation, increased mountain building in the Himalayan-Tibetan region, and decreased hydrothermal activity. None of the mechanisms alone seems to adequately explain the increased 87Sr/86Sr ratios during the Oligocene. Glaciation as a weathering agent was too episodic and probably began too late to explain the upturn in marine 87Sr/86Sr ratios. There is evidence that uplift in the Himalayan-Tibetan region began in the Miocene, much too late to control Oligocene strontium isotope ratios. Lastly, hydrothermal flux changes since the Eocene were apparently not great enough alone to account for the rise in marine 87Sr/86Sr ratios. We suggest that a combination of causes, such as decreased hydrothermal activity perhaps followed by increased glaciation and mountain building, might best explain the growth of the marine 87Sr/86Sr curve during the Oligocene.

Revised calibration of the MBT-CBT paleotemperature proxy

The MBT-CBT proxy for the reconstruction of paleotemperatures and past soil pH is based on the distribution of branched glycerol dialkyl glycerol tetraether (brGDGT) membrane lipids. The Methylation of Branched Tetraether (MBT) and the Cyclisation of Branched Tetraether (CBT) indices were developed to quantify these distributions, and significant empirical relations between these indices and annual mean air temperature (MAT) and/or soil pH were found in a large data set of soils. In this study, we extended this soil dataset to 278 globally distributed surface soils. Of these soils, 26% contains all nine brGDGTs, while in 63% of the soils the seven most common brGDGTs were detected, and the latter were selected for calibration purposes. This resulted in new transfer functions for the reconstruction of pH based on the CBT index: pH = 7.90-1.97 × CBT (r**2 = 0.70; RMSE = 0.8; n = 176), as well as for MAT based on the CBT index and methylation index based on the seven most abundant GDGTs (defined as MBT'): MAT = 0.81-5.67 × CBT + 31.0 × MBT' (r**2 = 0.59; RMSE = 5.0 °C; n = 176). The new transfer function for MAT has a substantially lower correlation coefficient than the original equation (r**2 = 0.77). To investigate possible improvement of the correlation, we used our extended global surface soil dataset to statistically derive the indices that best describe the relations of brGDGT composition with MAT and soil pH. These new indices, however, resulted in only a relatively minor increase in correlation coefficients, while they cannot be explained straightforwardly by physiological mechanisms. The large scatter in the calibration cannot be fully explained by local factors or by seasonality, but MAT for soils from arid regions are generally substantially (up to 20 °C) underestimated, suggesting that absolute brGDGT-based temperature records for these areas should be interpreted with caution. The applicability of the new MBT'-CBT calibration function was tested using previously published MBT-CBT-derived paleotemperature records covering the last deglaciation in Central Africa and East Asia, the Eocene-Oligocene boundary and the Paleocene-Eocene thermal maximum. The results show that trends remain similar in all records, but that absolute temperature estimates and the amplitude of temperature changes are lower for most records, and generally in better agreement with independent proxy data.

Radiocarbon age determination of 16 sites

The thermal diffusion enrichment apparatus in use in Amsterdam before 1967, has been rebuilt in the Groningen Radiocarbon Dating Laboratory. It has been shown to operate reliably and reproducibly. A reasonable agreement exists between the theoretical calculations and the experimental results. The 14C enrichment of a CO sample is deduced from the simultaneous mass 30 enrichment, which is measured with a mass spectrometer. The relation between both enrichments follows from a series of calibration measurements. The over-all accuracy in the enrichment is a few percent, equivalent to a few hundred years in age. The main problem in dating very old samples is their possible contamination with recent carbon. Generally, careful sample selection and rigorous pretreatment reduce sample contamination to an acceptable value. Also, it has been established that laboratory contamination, due to a memory effect in the combustion system and to impurities in the oxygen and nitrogen gas used for combustion, can be eliminated. A detailed analysis shows that the counter background in our set-up is almost exclusively caused by cosmic ray muons. The measurement of 28 early glacial samples, mostly from North-west Europe, has yielded a consistent set of ages. These indicate the existence of three early glacial interstadials; using the Weichselian definitions: Amersfoort starting at 68 200 ± 1100, Brørup at 64 400 ± 800 and Odderade at 60 500 ± 600 years BP. This 14C chronology shows good agreement with the Camp Century chronology and the dated palaeo sea levels. The discrepancy in the age of the early part of the Last Glacial on the 14C time scale and on that adopted for the deep-sea d18 record, must probably be attributed to the use of a generalized d18 curve and a wrong interpretation of this curve in terms of three Barbados terraces.