Tumors exhibit genomic and phenotypic heterogeneity, which has prognostic significance and may influence response to therapy. Imaging can quantify the spatial variation in architecture and function of individual tumors through quantifying basic biophysical parameters such as CT density or MRI signal relaxation rate; through measurements of blood flow, hypoxia, metabolism, cell death, and other phenotypic features; and through mapping the spatial distribution of biochemical pathways and cell signaling networks using PET, MRI, and other emerging molecular imaging techniques. These methods can establish whether one tumor is more or less heterogeneous than another and can identify subregions with differing biology. In this article, we review the image analysis methods currently used to quantify spatial heterogeneity within tumors. We discuss how analysis of intratumor heterogeneity can provide benefit over more simple biomarkers such as tumor size and average function. We consider how imaging methods can be integrated with genomic and pathology data, instead of being developed in isolation. Finally, we identify the challenges that must be overcome before measurements of intratumoral heterogeneity can be used routinely to guide patient care. Clin Cancer Res; 21(2); 249–57. ©2014 AACR.

Background: Drug-induced interstitial lung disease (DIILD) occurs as a result of numerous agents, but the risk often only becomes apparent after the marketing authorisation of such agents. Methods: In this PRISMA-compliant systematic review, we aimed to evaluate and synthesise the current literature on DIILD. Results: Following a quality assessment, 156 full-text papers describing more than 6000 DIILD cases were included in the review. However, the majority of the papers were of low or very low quality in relation to the review question (78%). Thus, it was not possible to perform a meta-analysis, and descriptive review was undertaken instead. DIILD incidence rates varied between 4.1 and 12.4 cases/million/year. DIILD accounted for 3–5% of prevalent ILD cases. Cancer drugs, followed by rheumatology drugs, amiodarone and antibiotics, were the most common causes of DIILD. The radiopathological phenotype of DIILD varied between and within agents, and no typical radiological pattern specific to DIILD was identified. Mortality rates of over 50% were reported in some studies. Severity at presentation was the most reliable predictor of mortality. Glucocorticoids (GCs) were commonly used to treat DIILD, but no prospective studies examined their effect on outcome. Conclusions: Overall high-quality evidence in DIILD is lacking, and the current review will inform larger prospective studies to investigate the diagnosis and management of DIILD.

We describe a method for automatically building statistical shape models from a training set of example boundaries/surfaces. These models show considerable promise as a basis for segmenting and interpreting images. One of the drawbacks of the approach is, however, the need to establish a set of dense correspondences between all members of a set of training shapes. Often this is achieved by locating a set of "landmarks" manually on each training image, which is time consuming and subjective in two dimensions and almost impossible in three dimensions. We describe how shape models can be built automatically by posing the correspondence problem as one of finding the parameterization for each shape in the training set. We select the set of parameterizations that build the "best" model. We define "best" as that which minimizes the description length of the training set, arguing that this leads to models with good compactness, specificity and generalization ability. We show how a set of shape parameterizations can be represented and manipulated in order to build a minimum description length model. Results are given for several different training sets of two-dimensional boundaries, showing that the proposed method constructs better models than other approaches including manual landmarking-the current gold standard. We also show that the method can be extended straightforwardly to three dimensions.

There is a clinical need for noninvasive biomarkers of tumor hypoxia for prognostic and predictive studies, radiotherapy planning, and therapy monitoring. Oxygen-enhanced MRI (OE-MRI) is an emerging imaging technique for quantifying the spatial distribution and extent of tumor oxygen delivery in vivo. In OE-MRI, the longitudinal relaxation rate of protons (ΔR₁) changes in proportion to the concentration of molecular oxygen dissolved in plasma or interstitial tissue fluid. Therefore, well-oxygenated tissues show positive ΔR₁. We hypothesized that the fraction of tumor tissue refractory to oxygen challenge (lack of positive ΔR₁, termed “Oxy-R fraction”) would be a robust biomarker of hypoxia in models with varying vascular and hypoxic features. Here, we demonstrate that OE-MRI signals are accurate, precise, and sensitive to changes in tumor pO₂ in highly vascular 786-0 renal cancer xenografts. Furthermore, we show that Oxy-R fraction can quantify the hypoxic fraction in multiple models with differing hypoxic and vascular phenotypes, when used in combination with measurements of tumor perfusion. Finally, Oxy-R fraction can detect dynamic changes in hypoxia induced by the vasomodulator agent hydralazine. In contrast, more conventional biomarkers of hypoxia (derived from blood oxygenation-level dependent MRI and dynamic contrast–enhanced MRI) did not relate to tumor hypoxia consistently. Our results show that the Oxy-R fraction accurately quantifies tumor hypoxia noninvasively and is immediately translatable to the clinic. Cancer Res; 76(4); 787–95. ©2015 AACR.