Vertical eDNA distribution of cold-water fishes in response to environmental variables in stratified lake

In summer, the survival zones of cold-water species are predicted to narrow by both increasing water temperatures from the surface and by expanding hypoxic zones from the lake bottom. To examine how the abundance of cold-water fishes changes along environmental gradients, we assessed the vertical environmental DNA (eDNA) distributions of three salmonid species which may have different water temperature tolerances during both stratification and turnover periods using quantitative PCR (qPCR). In addition, we examined on the vertical distribution of diverse fish fauna using an eDNA metabarcoding assay. The results suggested that the kokanee salmon (Oncorhynchus nerka) eDNA were abundant in deep, cold waters. On the other hand, rainbow trout (O. mykiss) eDNA were distributed uniformly throughout the water column, suggesting that they may have high water-temperature tolerance compared with kokanee salmon. The eDNA concentrations of masu salmon (O. masou) were below the detection limit (i.e., <10 copies μL-1) at all stations and depths and hence could not be quantified during stratification. Together with the finding that the eDNA distributions of other prey fish species were also constrained vertically in species-specific ways, our results suggest that climate change will result in substantial changes in the vertical distributions of lake fish species and thus affect their populations and interactions.

Can warming accelerate the decline of Odonata species in experimental paddies due to insecticide fipronil exposure?

Systemic insecticides are one of the causes of Odonata declines in paddy fields. Since rising temperatures associated with global warming can contribute to strengthen pesticide toxicity, insecticide exposures under increasing temperatures may accelerate the decline of Odonata species in the future. However, the combined effects of multiple stressors on Odonata diversity and abundance within ecosystems under various environmental conditions and species interactions are little known. Here, we evaluate the combined effects of the insecticide fipronil and warming on the abundance of Odonata nymphs in experimental paddies. We show that the stand-alone effect of the insecticide exposure caused a significant decrease in abundance of the Odonata community, while nymphs decreased synergistically in the combined treatments with temperature rise in paddy water. However, impacts of each stressor alone were different among species. This study provides experimental evidence that warming could accelerate a reduction in abundance of the Odonata community exposed to insecticides (synergistic effect), although the strength of that effect might vary with the community composition in targeted habitats, due mainly to different susceptibilities among species to each stressor. Community-based monitoring in actual fields is deemed necessary for a realistic evaluation of the combined effects of multiple stressors on biodiversity.

Noise pollution alters the diet composition of invertebrate consumers both in and beyond a noise-exposed grassland ecosystem

Anthropogenic noise is ubiquitous globally. However, we know little about how the impacts of noise alter fundamental ecosystem properties, such as resource consumption by invertebrate consumers. Using experimental noise manipulation and faecal DNA metabarcoding, we assessed how the direct and cross-trophic indirect effects of noise altered the dietary richness and specializations of omnivorous grasshoppers in a grassland ecosystem. We found that the experimental noise treatment expanded grasshoppers' dietary richness and resulted in dietary generalizations in both noise-exposed and adjacent relatively quieter areas. Unexpectedly, however, these dietary changes were primarily explained by the direct effect of noise not only in the noise-exposed areas but also in the adjacent quieter areas and were relaxed by indirect effects of noise such as reduced birds and predation risk and increased grasshoppers. Our work suggests that noise pollution can be key in explaining the variation of invertebrate consumers' diets across a gradient of noise-exposed environments.

Perspective: sustainability challenges, opportunities and solutions for long-term ecosystem observations

As interest in natural capital grows and society increasingly recognizes the value of biodiversity, we must discuss how ecosystem observations to detect changes in biodiversity can be sustained through collaboration across regions and sectors. However, there are many barriers to establishing and sustaining large-scale, fine-resolution ecosystem observations. First, comprehensive monitoring data on both biodiversity and possible anthropogenic factors are lacking. Second, some in situ ecosystem observations cannot be systematically established and maintained across locations. Third, equitable solutions across sectors and countries are needed to build a global network. Here, by examining individual cases and emerging frameworks, mainly from (but not limited to) Japan, we illustrate how ecological science relies on long-term data and how neglecting basic monitoring of our home planet further reduces our chances of overcoming the environmental crisis. We also discuss emerging techniques and opportunities, such as environmental DNA and citizen science as well as using the existing and forgotten sites of monitoring, that can help overcome some of the difficulties in establishing and sustaining ecosystem observations at a large scale with fine resolution. Overall, this paper presents a call to action for joint monitoring of biodiversity and anthropogenic factors, the systematic establishment and maintenance of in situ observations, and equitable solutions across sectors and countries to build a global network, beyond cultures, languages, and economic status. We hope that our proposed framework and the examples from Japan can serve as a starting point for further discussions and collaborations among stakeholders across multiple sectors of society. It is time to take the next step in detecting changes in socio-ecological systems, and if monitoring and observation can be made more equitable and feasible, they will play an even more important role in ensuring global sustainability for future generations. This article is part of the theme issue 'Detecting and attributing the causes of biodiversity change: needs, gaps and solutions'.

Assessing ecosystem vulnerability under severe uncertainty of global climate change

Assessing the vulnerability and adaptive capacity of species, communities, and ecosystems is essential for successful conservation. Climate change, however, induces extreme uncertainty in various pathways of assessments, which hampers robust decision-making for conservation. Here, we developed a framework that allows us to quantify the level of acceptable uncertainty as a metric of ecosystem robustness, considering the uncertainty due to climate change. Under the framework, utilizing a key concept from info-gap decision theory, vulnerability is measured as the inverse of maximum acceptable uncertainty to fulfill the minimum required goal for conservation. We applied the framework to 42 natural forest ecosystems and assessed their acceptable uncertainties in terms of maintenance of species richness and forest functional type. Based on best-guess estimate of future temperature in various GCM models and RCP scenarios, and assuming that tree species survival is primarily determined by mean annual temperature, we performed simulations with increasing deviation from the best-guess temperature. Our simulations indicated that the acceptable uncertainty varied greatly among the forest plots, presumably reflecting the distribution of ecological traits and niches among species within the communities. Our framework provides acceptable uncertainty as an operational metric of ecosystem robustness under uncertainty, while incorporating both system properties and socioeconomic conditions. We argue that our framework can enhance social consensus building and decision-making in the face of the extreme uncertainty induced by global climate change.

Inferring causal impacts of extreme water-level drawdowns on lake water clarity using long-term monitoring data

Although long-term ecosystem monitoring provides essential knowledge for practicing ecosystem management, analyses of the causal effects of ecological impacts from large-scale observational data are still in an early stage of development. We used causal impact analysis (CIA)-a synthetic control method that enables estimation of causal impacts from unrepeated, long-term observational data-to evaluate the causal impacts of extreme water-level drawdowns during summer on subsequent water quality. We used more than 100 years of transparency and water level monitoring data from Lake Biwa, Japan. The results of the CIA showed that the most extreme drawdown in recorded history, which occurred in 1994, had a significant positive effect on transparency (a maximum increase of 1.75 m on average over the following year) in the north basin of the lake. The extreme drawdown in 1939 was also shown to be a trigger for an increase in transparency in the north basin, whereas that in 1984 had no significant effects on transparency. In the south basin, contrary to the pattern in the north basin, the extreme drawdown had a significant negative effect on transparency shortly after the extreme drawdown. These different impacts of the extreme drawdowns were considered to be affected by the timing and magnitude of the extreme drawdowns and the depths of the basins. Our approach of inferring the causal impacts of past events on ecosystems will be helpful in implementing water-level management for ecosystem management and improving water quality in lakes.

Narrowly distributed taxa are disproportionately informative for conservation planning

Biological atlas data can be used as inputs into conservation decision-making, yet atlases are sometimes infrequently updated, which can be problematic when the distribution of species is changing rapidly. Despite this, we have a poor understanding of strategies for efficiently updating biological atlas data. Using atlases of the distributions of 1630 threatened plant taxa, we quantitatively compared the informativeness of narrowly distributed and widespread taxa in identifying areas that meet taxon-specific conservation targets, and also measured the cost-efficiency of meeting those targets. We also explored the underlying mechanisms of the informativeness of narrowly distributed taxa. Overall, narrowly distributed taxa are far more informative than widespread taxa for identifying areas that efficiently meet conservation targets, while their informativeness for identifying cost-efficient areas varied depending on the type of conservation target. Narrowly distributed taxa are informative mainly because their distributions disproportionately capture areas that are either relatively taxon rich or taxon poor, and because of larger number of taxa captured with given number of records. Where resources for updating biological data are limited, a focus on areas supporting many narrowly distributed taxa could benefit conservation planning.

Complex range shifts among forest functional types under the contemporary warming

The direction and magnitude of species distribution shifts tend to differ among species and functional types (FTs). Quantifying functional trait variation and species interactions will improve our understanding of the complex mechanisms that govern ecosystem dynamics and their responses to climate change. Here, we analyzed differences in the juvenile and adult temperature ranges of Japanese tree species at the mean, colder edge, and warmer edge of their distributions to reveal how functional traits affect interactions between different FT groups (e.g., deciduous and evergreen broad-leaved trees), using linear models and permutation tests. Overall, juveniles preferred cooler sites, but with high variation. The variation among species was partly explained by the difference in seed mass where species with lighter seeds tend to colonize colder sites. On the other hand, the distribution range of FTs showed complex behavior at the ecotones of different FTs. Specifically, in three of eight ecotones, nonparallel range shifts between FTs were detected, which includes cold shifting in deciduous broad-leaved FT where a warm shift by subalpine FT happened, and cold shifting in subtropical FT where warm shifts by either the deciduous broad-leaved or the evergreen broad-leaved FTs happened. Our results suggest that past warming has caused a general cold shift at species level, whereas different mechanisms, such as light seeds disperse farther in distribution's colder edge and heavy seeds (e.g., evergreen broad-leaved) compete better in warmer edge, create nonparallel responses of FT distribution ranges leading to the observed homogenization at several ecotones among FTs. These complex range shifts at FT level have crucial implications for climate change mitigation and adaptation.

Emergent dual scaling of riverine biodiversity

A prevailing paradigm suggests that species richness increases with area in a decelerating way. This ubiquitous power law scaling, the species-area relationship, has formed the foundation of many conservation strategies. In spatially complex ecosystems, however, the area may not be the sole dimension to scale biodiversity patterns because the scale-invariant complexity of fractal ecosystem structure may drive ecological dynamics in space. Here, we use theory and analysis of extensive fish community data from two distinct geographic regions to show that riverine biodiversity follows a robust scaling law along the two orthogonal dimensions of ecosystem size and complexity (i.e., the dual scaling law). In river networks, the recurrent merging of various tributaries forms fractal branching systems, where the prevalence of branching (ecosystem complexity) represents a macroscale control of the ecosystem's habitat heterogeneity. In the meantime, ecosystem size dictates metacommunity size and total habitat diversity, two factors regulating biodiversity in nature. Our theory predicted that, regardless of simulated species' traits, larger and more branched "complex" networks support greater species richness due to increased space and environmental heterogeneity. The relationships were linear on logarithmic axes, indicating power law scaling by ecosystem size and complexity. In support of this theoretical prediction, the power laws have consistently emerged in riverine fish communities across the study regions (Hokkaido Island in Japan and the midwestern United States) despite hosting different fauna with distinct evolutionary histories. The emergence of dual scaling law may be a pervasive property of branching networks with important implications for biodiversity conservation.

Current site planning of medium to large solar power systems accelerates the loss of the remaining semi-natural and agricultural habitats

The global transition to renewable energy sources has accelerated to mitigate the effects of global climate change. Sudden increases in solar power facilities have caused the physical destruction of wildlife habitats, thereby resulting in the decline of biodiversity and ecosystem functions. However, previous assessments have been based on the environmental impact of large solar photovoltaics (PVs). The impact of medium-sized PV facilities (0.5-10 MW), which can alter small habitat patches through the accumulation of installations has not been assessed. Here, we quantified the amount of habitat loss directly related to the construction of PV facilities with different size classes and estimated their siting attributes using construction patterns in Japan and South Korea. We identified that a comparable amount of natural and semi-natural habitats were lost due to the recent installation of medium solar facilities (approximately 66.36 and 85.73% of the overall loss in Japan and South Korea, respectively). Compared to large solar PVs, medium PV installations resulted in a higher area loss of semi-natural habitats, including secondary/planted forests, secondary/artificial grasslands, and agricultural lands. The siting attributes of medium and large solar PV facilities indicated a preference for cost-based site selection rather than prioritizing habitat protection for biodiversity conservation. Moreover, even conservation areas were developed when economic and topological conditions were suitable for energy production. Our simulations indicate that increasing the construction of PVs in urban areas could help reduce the loss of natural and semi-natural habitats. To improve the renewable energy share while mitigating the impacts on biodiversity, our results stress the need for a proactive assessment to enforce sustainable site-selection criteria for solar PVs in renewable energy initiatives. The revised criteria should consider the cumulative impacts of varied size classes of solar power facilities, including medium PVs, and the diverse aspects of the ecological value of natural habitats.

Forest Tax Payment Responsibility from the Forest Service Footprint Perspective

It has been observed that market failure has hampered the development of sustainable forest ecosystem services such as CO₂ absorption and fixation, water retention, and biodiversity. One of the reasons for this is that the link between forest land use and the beneficiaries of that use has not been widely recognized or clearly established. To address this problem, we conducted a footprint analysis to clarify the linkage between Japanese taxpayers as the beneficiaries of forest land use and the use of tax revenue and monetary donations for forest management. This study focuses on how the current forest tax collected from Japanese taxpayers (63 billion Japanese yen) could be allocated more fairly. The question of whether the collected taxes are sufficient is left for another time. At the core of our analysis, we examined the carbon footprint and established a linkage between the origins of CO₂ emissions in Japan and their destinations by using a subnational multiregional input-output database and building a base table focused on various land use types and subnational regions at the municipality level. By clarifying these linkages and enhancing their transparency, we provide a basis for developing alternative financing schemes involving both taxation and taxpayer donations in support of forest management activities and protection of biodiverse habitats.

Development of a camera trap for perching dragonflies: a new tool for freshwater environmental assessment

Although dragonflies are excellent environmental indicators for monitoring terrestrial water ecosystems, automatic monitoring techniques using digital tools are limited. We designed a novel camera trapping system with an original dragonfly detector based on the hypothesis that perching dragonflies can be automatically detected using inexpensive and energy-saving photosensors built in a perch-like structure. A trial version of the camera trap was developed and evaluated in a case study targeting red dragonflies (Sympetrum spp.) in Japan. During an approximately 2-month period, the detector successfully detected Sympetrum dragonflies while using extremely low power consumption (less than 5 mW). Furthermore, a short-term field experiment using time-lapse cameras for validation at three locations indicated that the detection accuracy was sufficient for practical applications. The frequency of false positive detection ranged from 17 to 51 over an approximately 2-day period. The detection sensitivities were 0.67 and 1.0 at two locations, where a time-lapse camera confirmed that Sympetrum dragonflies perched on the trap more than once. However, the correspondence between the detection frequency by the camera trap and the abundance of Sympetrum dragonflies determined by field observations conducted in parallel was low when the dragonfly density was relatively high. Despite the potential for improvements in our camera trap and its application to the quantitative monitoring of dragonflies, the low cost and low power consumption of the detector make it a promising tool.

Investigating effect of climate warming on the population declines of Sympetrum frequens during the 1990s in three regions in Japan

Climate warming is of concern as a key factor in the worldwide decline in insect populations. In Japan, numbers of a common dragonfly in rice paddy fields, Sympetrum frequens, decreased sharply in the 1990s. Because S. frequens migrates to cooler mountains in summer, climate warming has been suggested as one of the main causes of the population decline in addition to agronomic factors. Here, we analysed the relation between summer temperatures and population densities of S. frequens and the related S. infuscatum, which does not migrate to mountains in summer, using published population monitoring data and temperature data from three regions (Toyama, Ishikawa, and Shizuoka) in Japan. Decadal differences in summer temperatures lay within the range of fluctuations among years, suggesting that an increase in summer temperatures cannot explain the past sharp population declines. However, regression analyses using monitoring data from Toyama showed that the population dynamics of both species in autumn are negatively correlated with summer temperatures in the same year. These results suggest that high temperatures in summer directly affect adult mortality to an extent that results in a decrease in population growth.

Direct and indirect effects of noise pollution alter biological communities in and near noise-exposed environments

Noise pollution is pervasive across every ecosystem on Earth. Although decades of research have documented a variety of negative impacts of noise to organisms, key gaps remain, such as how noise affects different taxa within a biological community and how effects of noise propagate across space. We experimentally applied traffic noise pollution to multiple roadless areas and quantified the impacts of noise on birds, grasshoppers and odonates. We show that acoustically oriented birds have reduced species richness and abundance and different community compositions in experimentally noise-exposed areas relative to comparable quiet locations. We also found both acoustically oriented grasshoppers and odonates without acoustic receptors to have reduced species richness and/or abundance in relatively quiet areas that abut noise-exposed areas. These results suggest that noise pollution not only affects acoustically oriented animals, but that noise may reverberate through biological communities through indirect effects to those with no clear links to the acoustic realm, even in adjacent quiet environments.

Homogenization of freshwater lakes: Recent compositional shifts in fish communities are explained by gamefish movement and not climate change

Globally, lake fish communities are being subjected to a range of scale-dependent anthropogenic pressures, from climate change to eutrophication, and from overexploitation to species introductions. As a consequence, the composition of these communities is being reshuffled, in most cases leading to a surge in taxonomic similarity at the regional scale termed homogenization. The drivers of homogenization remain unclear, which may be a reflection of interactions between various environmental changes. In this study, we investigate two potential drivers of the recent changes in the composition of freshwater fish communities: recreational fishing and climate change. Our results, derived from 524 lakes of Ontario, Canada sampled in two periods (1965-1982 and 2008-2012), demonstrate that the main contributors to homogenization are the dispersal of gamefish species, most of which are large predators. Alternative explanations relating to lake habitat (e.g., area, phosphorus) or variations in climate have limited explanatory power. Our analysis suggests that human-assisted migration is the primary driver of the observed compositional shifts, homogenizing freshwater fish community among Ontario lakes and generating food webs dominated by gamefish species.

Abstract P6-02-11: The TILs-US scores based on ultrasonography can predict lymphocyte-predominant breast cancer before surgery

[Purpose]

Tumor-infiltrating lymphocytes (TILs) has been shown to be useful for predicting outcomes after surgery in breast cancer, and while TILs can be evaluated in preoperative biopsy tissue, heterogeneous distribution of TILs requires examination of all biopsied tissue samples.

We gave scores to preoperative ultrasonography (US) images with characteristics indicative of lymphocyte-predominant breast cancer (LPBC) and attempted to apply these for diagnostic prediction of LPBC. In this study, TILs-US scores based on preoperative US were assessed for their usefulness in predicting LPBC, the diagnosis of which was confirmed with postoperative pathology.

[Methods]

We evaluated 161 patients with invasive breast cancer between January 2014 and December 2017. All patients were treated by either mastectomy or breast-conserving surgery. Stromal lymphocytes were evaluated on preoperative biopsy tissues and surgical pathological specimens. Breast cancer samples with ≥ 50% stromal TILs were defined as pre-LPBC (preoperative biopsy tissues) and LPBC (surgical pathological specimens). TILs-US score was calculated from US before curative surgery. Based on clinicopathological factors including TILs-US scores based on preoperative US and pre-LPBC indicators, determinants useful for prediction of LPBC were examined.

[Results]

There were 39 cases of LPBCs and 122 cases of non-LPBCs in surgical pathological specimens. We set TILs-US score cut-offs for predicting LPBC at 4 points based on the receiver operating characteristics (ROC) curves (AUC, 0.88). There were significant predictors for LPBC in maltivariate logistic analysis (TILs-US score: OR26.8, p&lt;0.001; pre- LPBC: 18.6, p=0.002; HER-2: OR9.2, p=0.009) in preoperative clinicopathological factor. The sensitivity, specificity and accuracy of TILs-US score for predicting LPBC were 0.74 (0.62-0.84), 0.89 (0.85-0.92) and 0.85 (0.79-0.90). Those of pre-LPBC were 0.51(0.42-0.55), 0.98 (0.96-1.00) and 0.87 (0.82-0.89), and those of HER2 were 0.28(0.19-0.36), 0.94(0.91-0.97) and 0.78 (0.74-0.82), respectively. The sensitivity of TILs-US score for predicting LPBC was significantly greater than those of pre- LPBC (p=0.04) and HER2 (p&lt;0.001). On the other hand, the specificity of pre- LPBC for predicting LPBC was significantly greater than that of TILs-US score(p=0.002).

The sensitivity, specificity and accuracy of predicting LPBCPerformance measureTILs-US score (95%CI)Preoperative biopsyHER-2P, TILs-US score vs. Preoperative biopsyP, TILs-USscore vs. HER-2Sensitivity0.74 (0.62-0.84)0.51 (0.42-0.55)0.28 (0.19-0.36)0.04&lt;0.001Specificity0.89 (0.85-0.92)0.98 (0.96-1.00)0.94 (0.91-0.97)0.0020.11Accuracy0.85 (0.79-0.90)0.87 (0.82-0.89)0.78 (0.74-0.82)0.230.11

[Conclusions]

TILs-US scores are an important factor that can predict LPBC preoperatively. The TILs-US score has particularly high sensitivity and may be an applicable index in the preoperative evaluation for LPBC.

Citation Format: Kanou A, Masumoto N, Shiroma N, Fukui K, Sasada S, Emi A, Kadoya T, Yokozaki M, Arihiro K, Okada M. The TILs-US scores based on ultrasonography can predict lymphocyte-predominant breast cancer before surgery [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P6-02-11.

Abstract P6-02-06: Characteristics of lymphocyte-predominant breast cancer in ultrasound images and their application to diagnostic prediction

[Purpose]

Tumor-infiltrating lymphocytes (TILs) is a prognostic factor for breast cancer, however, an accurate and simple evaluation method remains elusive. Therefore, we focused on findings characteristic of lymphocyte-predominant breast cancer (LPBC) in ultrasound (US) images. In this study, the application of preoperative US image assessment to diagnostic prediction of LPBC evaluated from postoperative pathological specimens, was appraised.

[Methods]

We evaluated 191 patients with invasive breast cancer between January 2014 and December 2017. All patients were treated by either mastectomy or breast-conserving surgery. Stromal lymphocytes were evaluated on surgical pathological specimens. Breast cancer samples with ≥ 50% stromal TILs were defined as LPBC. Preoperative US was performed in all cases and images were examined for characteristics indicative of TILs. Scores were given to US images with characteristic TILs and these TILs-US scores were assessed for their application to predict LPBC.

[Results]

There were 39 cases of LPBCs and 122 cases of non-LPBCs in surgical pathological specimens.

The characteristic US image findings predicting LPBC were shape (more lobulated), internal echo level (weaker) and posterior echoes (stronger). The TILs-US scores were given based on these three ultrasound tissue characterizations.We set TILs-US score cut-offs for predicting LPBC at 4 points (Sensitivity, 0.73; specificity, 0.87; accuracy, 0.83) based on the receiver operating characteristics (ROC) curves (AUC, 0.88). There were significant predictors for LPBC in maltivariate logistic analysis (Nuclear Grade (NG): OR3.4, p=0.02; ER: 5.7, p =0.007;HER-2: OR4.1, p=0.04; TILs-US score2: OR14.9, p&lt;0.001) in preoperative clinicopathological factor. The sensitivity, specificity and accuracy of NG for predicting LPBC were 0.75, 0.69 and 0.71. Those of ER and HER2 were 0.33, 0.96 and 0.79.Sensitivity, specificity, and accuracy of NG, ER, and HER2 diagnoses were all lower than the TILs-US score, and the TILs-US score showed the best diagnostic ability.

The sensitivity, specificity and accuracy of predicting LPBCPerformance measureTILs-US score (95%CI)NGER or HER-2P, TILs-US score vs. NG,P, TILs-USscore vs. ER or HER-2Sensitivity0.73 (0.63-0.81)0.75 (0.64-0.84)0.327 (0.24-0.39)0.08&lt;0.001specificity0.87 (0.83-0.90)0.69 (0.65-0.72)0.957 (0.93-0.98)&lt;0.001&lt;0.001accuracy0.83 (0.77-0.88)0.707 (0.65-0.376)0.785 (0.74-0.82)0.0040.25

Conclusions

LPBC has characteristic ultrasound tissue characterizations in US images. Thus, TILs-US scores based on US may be applicable to accurate and convenient preoperative diagnosis of LPBC.

Citation Format: Fukui K, Masumoto N, Shiroma N, Kanou A, Yokozaki M, Sasada S, Emi A, Kadoya T, Arihiro K, Okada M. Characteristics of lymphocyte-predominant breast cancer in ultrasound images and their application to diagnostic prediction [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P6-02-06.

Abstract P6-02-09: Diagnostic performance of dedicated breast PET for the prediction of pathological response after neoadjuvant chemotherapy

Background: Neoadjuvant chemotherapy (NAC) is a standard treatment for operable breast cancer. However, imaging methods for evaluating treatment response have not been established. Previous studies reported that ring-type dedicated breast positron emission tomography (DbPET) detected residual tumors following NAC more accurately than whole-body PET/CT. This study assessed DbPET parameters for predicting pathological complete response (pCR) in patients with breast cancer.

Patients and Methods: Among patients with breast cancer who underwent surgery after NAC, 61 were examined using ring-type DbPET before and after NAC. The maximum standardized uptake values (SUVmax) and tumor-to-normal-tissue ratio (TNR) were calculated before and after NAC (pre-SUVmax, pre-TNR, post-SUVmax and post-TNR, respectively). Moreover, the reduction rates (ΔSUVmax and ΔTNR) were determined. pCR was defined as complete remission of breast cancer.

Results: The median patient age was 52 years. Forty patients (65.6%) were estrogen receptor (ER)-positive, whereas 25 patients (41.0%) were HER2-positive. Fifteen patients (24.6%) achieved pCR after NAC. The calculated values for the parameters of DbPET are summarized in Table 1. The most promising parameters for predicting pCR were ΔSUVmax (area under the curve [AUC]: 0.506) and post-TNR (AUC: 0.640). Although neither of these two parameters reflected the pathological response to NAC in patients with ER-positive disease, post-TNR showed the highest AUC (i.e., AUC: 0.750) for pCR in patients with ER-negative disease. The sensitivity and specificity of post-TNR in the ER-positive group were 85.7% and 39.4%, respectively. In the ER-negative group, these values were 100% and 58.3%, respectively [Table 2]

Table 1.Diagnostic performance of dedicated breast positron-emission tomography for the prediction of pathological complete response after neoadjuvant chemotherapyParameterspCR Median (IQR)Non-pCR Median (IQR)PAUC (95% CI)Pre-SUVmax14.8 (9.2–17.7)14.4 (10.7–19.8)0.6040.454 (0.285–0.624)Post-SUVmax1.8(1.5–2.1)2.0 (1.6–3.4)0.2370.603 (0.460–0.746)ΔSUVmax (%)87.32 (81.5–91.8)88.6(82.8–91.8)0.9540.506 (0.329–0.682)Pre-TNR8.0 (5.3–9.2)7.3(5.9–10.2)0.7570.472 (0.297–0.647)Post-TNR1.7(1.0–1.1)1.2 (1.0–2.4)0.0980.640 (0.506–0.774)ΔTNR (%)87.3 (78.3–89.1)79.8 (62.7–86.7)0.1130.638 (0.472–0.803)pCR, pathological complete response; IQR, interquartile range; AUC, area under the curve; CI, confidence interval, SUV, standardized uptake value; TNR, tumor-to-normal-tissue ratio.Table 2.Diagnostic accuracy of post-TNR according to estrogen receptor status Sensitivity (%)Specificity (%)Accuracy (%)PPV (%)NPV (%)ER-positive85.739.447.523.192.9ER-negative10058.375.061.5100TNR, tumor-to-normal-tissue ratio; ER, estrogen receptor; PPV, positive predictive value; NPV, negative predictive value

Conclusion: In DbPET, ΔSUVmax and post-TNR were shown to be promising parameters for predicting pathological response to NAC. Post-TNR provided the highest sensitivity for predicting pCR in patients with ER-negative breast cancer.

Citation Format: Suzuki E, Sasada S, Sueoka S, Masumoto N, Goda N, Kajitani K, Emi A, Haruta R, Kadoya T, Kataoka T, Okada M. Diagnostic performance of dedicated breast PET for the prediction of pathological response after neoadjuvant chemotherapy [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P6-02-09.

Abstract OT2-07-05: A phase III trial to compare eribulin mesylate + trastuzumab (H) + pertuzumab (P) with paclitaxel or docetaxel + HP for HER2-positive advanced or metastatic breast cancer (JBCRG-M06/ EMERALD)

Background: Docetaxel + Trastuzumab (H) + Pertuzumab (P) provided progression-free survival (PFS) and overall survival (OS) benefits in HER2-positive advanced or metastatic breast cancer (AMBC) in the CLEOPATRA study as a first-line therapy. However, long-term administration of docetaxel at a dose of 75 mg/m2 every 3 weeks in AMBC patients (pts) is difficult due to the toxicities. Eribulin mesylate (E) is a well-tolerated microtubule inhibitor, and we have reported the efficacy and safety of EHP regimen as first- and second-line therapy for AMBC in a multicenter, phase II study (JBCRG-M03/UMIN000012232). In this M06 study, we address the clinical question as to which is the better chemotherapy partner for HP as first line regimen, in terms of efficacy, toxicity and QOL.

Methods: JBCRG-M06 is a multicenter open-label randomized phase III study for HER2-positive AMBC pts who have received no prior chemotherapy except for the HER2- Antibody-Drug Conjugate (ADC). Pts will be randomized 1:1 to E (1.4mg/m2 on day 1 and 8) + H (8 mg/kg loading dose followed by 6 mg/kg) +P (840 mg loading dose followed by 420 mg) q3wks or standard taxanes (docetaxel 75mg/m2 on day1 or paclitaxel 80mg/m2 on day 1, 8 and 15) + HP q3wks. Stratification factors for randomization are; presence of visceral metastases, number of prior taxanes on perioperative adjuvant treatment, and treatment with prior anti-HER2-ADC. Primary endpoint is PFS and secondary endpoints include overall response rate, duration of response, OS, patient-reported outcomes (PRO) relating to QOL and peripheral neuropathy, new-metastases free survival, and safety. Translational research to search for biomarker for individual precision therapy will be performed. Main eligibility criteria are as follows: pts with HER2-positive AMBC, female aged 20-70 years old, ECOG PS of 0-1, LVEF ≥ 50% at baseline and adequate organ function. Pts who had progressive MBC within 6 months after the end of primary adjuvant systemic chemotherapy are excluded. The sample size was calculated by type1 error (2-sided) of 0.05 and 80% power to estimate the noninferiority margin 1.33 with an expected median PFS of 14.2 months. The target number of pts is 480 recruited over the duration of 3-years. The first patient in was achieved on August 2017. (ClinicalTrials.gov Identifier:NCT03264547).

Citation Format: Masuda N, Yamashita T, Saji S, Araki K, Ito Y, Takano T, Takahashi M, Tsurutani J, Koizumi K, Kitada M, Kojima Y, Sagara Y, Tada H, Iwasa T, Kadoya T, Iwatani T, Hasegawa H, Morita S, Ohno S. A phase III trial to compare eribulin mesylate + trastuzumab (H) + pertuzumab (P) with paclitaxel or docetaxel + HP for HER2-positive advanced or metastatic breast cancer (JBCRG-M06/ EMERALD) [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr OT2-07-05.

Abstract P6-02-17: Hand-held impulse-radar detector for breast cancer: development and a pilot study

Background: Microwave breast imaging, which using the difference in the dielectric properties between breast cancer and normal breast tissue, is a painless and non-radiation method. We have created a novel hand-held prototype of breast cancer detector using impulse-radar based imaging system, and conducted a pilot clinical study.

Methods: The detector consists of complementary metal-oxide-semiconductor (CMOS) integrated circuits covering the ultrawideband width from 3.1 to 10.6 GHz, which enable the generation and transmission of Gaussian monocycle pulse (GMP) trains and single port eight throw switching matrices (SP8T-SW) for controlling a 4×4 cross-shaped dome antenna array. The size of the detector was 19.1 × 17.7 × 18.8 cm. After evaluation using a breast tumor phantom and the resected breast specimens obtained through mastectomy, we recruited 5 patients with histologically confirmed breast cancers in the clinical study. The detector was placed on the breast with the patient in a supine position. The primary endpoint was a detection rate of breast cancers, and the secondary endpoints were positional accuracy and adverse event. This study was registered with the UMIN Clinical Trials Registry (UMIN000026181).

Results: The three-dimensional positions of the tumors in the imaging results using a phantom and resected specimens are consistent with the results of histopathology analysis. In the clinical study, all 5 targeted breast cancers were detected and were visualized at the sites confirmed by other diagnostic modalities. Among 5 tumors, one was not detected via mammography because of heterogeneously dense breast and another was a microinvasive carcinoma of invasive tumor size 0.5 mm. No study-related adverse events occurred.

Conclusions: We succeeded in creating a new device of hand-held impulse-radar detector for breast cancer. The detector has sufficient detective capability, is safe for clinical use, and might detect an early stage breast cancer. In the future, we will proceed with the development to clinical application.

Citation Format: Sasada S, Masumoto N, Song H, Goda N, Kajitani K, Emi A, Kadoya T, Arihiro K, Kikkawa T, Okada M. Hand-held impulse-radar detector for breast cancer: development and a pilot study [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P6-02-17.

Consumer trophic positions respond variably to seasonally fluctuating environments

The effects of environmental seasonality on food web structure have been notoriously understudied in empirical ecology. Here, we focus on seasonal changes in one key attribute of a food web, consumer trophic position. We ask whether fishes inhabiting tropical river-floodplain ecosystems behave as seasonal omnivores, by shifting their trophic positions in relation to the annual flood pulse, or whether they feed at the same trophic position all year, as much empirical work implicitly assumes. Using dietary data from the Tonle Sap Lake, Cambodia, and a literature review, we find evidence that some fishes, especially small piscivores, increased consumption of invertebrates and/or plant material during the wet season, as predicted. However, nitrogen stable isotope (δ¹⁵ N) data for 26 Tonle Sap fishes, spanning a broader range of functional groups, uncovered high variation in seasonal trophic position responses among species (0 to ±0.52 trophic positions). Based on these findings, species respond to the flood pulse differently. Diverse behavioral responses to seasonality, underpinned by spatiotemporal variation at multiple scales, could be central for rerouting matter and energy flow in these dynamic ecosystems. Seasonally flexible foraging behaviors warrant further study given their potential influence on food web dynamics in a range of fluctuating environments.

Potential oscillators and keystone modules in food webs

Food web theory suggests that the placement of a weak interaction is critical such that under some conditions even one well-placed weak interaction can stabilise multiple strong interactions. This theory suggests that complex stable webs may be built from pivotal weak interactions such that the removal of even one to a few keystone interactions can have significant cascading impacts on whole system diversity and structure. However, the connection between weak interactions, derived from the theory of modular food web components, and keystone species, derived from empirical results, is not yet well understood. Here, we develop numerical techniques to detect potential oscillators hidden in complex food webs, and show that, both in random and real food webs, keystone consumer-resource interactions often operate to stabilise them. Alarmingly, this result suggests that nature frequently may be dangerously close to precipitous change with even the loss of one or a few weakly interacting species.

Context-dependent interactions and the regulation of species richness in freshwater fish

Species richness is regulated by a complex network of scale-dependent processes. This complexity can obscure the influence of limiting species interactions, making it difficult to determine if abiotic or biotic drivers are more predominant regulators of richness. Using integrative modeling of freshwater fish richness from 721 lakes along an 11^o latitudinal gradient, we find negative interactions to be a relatively minor independent predictor of species richness in lakes despite the widespread presence of predators. Instead, interaction effects, when detectable among major functional groups and 231 species pairs, were strong, often positive, but contextually dependent on environment. These results are consistent with the idea that negative interactions internally structure lake communities but do not consistently ‘scale-up’ to regulate richness independently of the environment. The importance of environment for interaction outcomes and its role in the regulation of species richness highlights the potential sensitivity of fish communities to the environmental changes affecting lakes globally.

Species richness patterns are driven by biotic and abiotic factors, the relative strengths of which are unclear. Here, the authors test how species interactions or environmental traits influence fish richness across over 700 Canadian lakes, showing a surprisingly small role of negative interactions.

Abstract P5-02-03: Evaluation of pathological malignancy grade and neoplastic progress of breast cancer using dedicated breast positron emission tomography

Background: Dedicated breast positron emission tomography (DbPET)provides detailed high resolution images of the breast and enables quantitative assessment using standard uptake values (SUVs). We aimed to determine whether DbPET can predict the pathological malignancy grade and neoplastic progress of breast cancer compared with whole body (WB) PET.

Methods: We investigated 196 consecutive patients with invasive breast cancerwho underwent concurrent Db- and WB-PET from January 2016 to March 2017. All Db- and WB-PET were quantified based on SUVs. We also investigated pathological features of breast cancer who had a ring-like uptake (RU) without central FDG accumulation on DbPET.

Results:

The associations between the SUVs for DB- and WB-PET and the pathological factors in breast cancerCharacteristicSUV WBPETDbPET Mean ± SDpMean ± SDpall3.6 ± 3.4 9.4±7.9 Tumor size    ≤2.0 cm2.2±1.6&lt;0.0016.5±5.2&lt;0.001&gt;2.0 cm5.5±4.1 13.3±9.2 LN    Negative3.1±3.3&lt;0.0018.4±7.8&lt;0.001Positive4.9±3.2 11.8±7.7 NG    1 or 22.4±2.1&lt;0.0016.6±5.9&lt;0.00135.1±4.0 12.7±8.7 Ki67    &lt; 201.8±1.1&lt;0.0015.2±3.3&lt;0.001≥ 204.4±3.7 11.4±8.6 ER    positive3.4±3.30.028.8±7.60.006negative5.2±3.6 13.5±8.6 HER-2    positive4.6±3.10.0911.8±7.60.04negative3.4±3.4 9.0±7.9 Sub type vs Lumnal A vs Lumnal ALuminal A1.8±1.1 5.2±3.3 Luminal B4.0±3.8&lt;0.00110.1±8.5&lt;0.001HER24.6±3.1&lt;0.00111.8±7.6&lt;0.001Triple negative5.3±3.8&lt;0.00113.8±9.2&lt;0.001

summarizes the association between SUVs for Db- and WB-PET and pathological factors inbreast cancer.SUVs on PET were significantly higher for the tumor size of &gt;2.0 cm than for tumor size ≤2.0 cm (p&lt;0.001), for LN-positive than for LN-negative (p&lt;0.001), for NG3 than for NG1-NG2 (p&lt;0.001), for higher Ki67 expression than for lower Ki67 expression (p&lt;0.001), and for ER-negative than for ER-positive (WBPET, p=0.02; DbPET, p=0.006). SUVs were significantly lower for Luminal A than for Luminal B, HER2, and triple-negative cancer (p&lt;0.001 for all three).SUVs for DbPET was significantly higher for HER2-positive than for HER2-negative (p=0.02).

The association between SUVs for breast cancer with and without RU on DbPETCharacteristicRU(-), nRU(+), npall17323 Tumor size   ≤2.0 cm1095&lt;0.001&gt;2.0 cm6418 LN   Negative1299&lt;0.001Positive4414 NG   1 or 210070.0237316 Ki67   &lt; 206130.03≥ 2011220 ER   positive152190.49negative214 HER-2   positive2620.38negative14721 Sub type  vs Lumnal ALuminal A493 Luminal B84150.02HER22620.81Triple negative1430.04

summarizes the association between SUVs for breast cancer with and without RU on DbPET. SUVs for breast cancer with RU on DbPET were significantly higher for the tumor size of &gt;2.0 cm than for tumor size ≤2.0 cm, for LN -positive than for LN-negative (p&lt;0.001), for NG3 than for NG1-2 (p=0.02), and for higher Ki67 expression than for lowerKi67 expression (p=0.03). SUVs were significantly lower for Luminal A than for Luminal B (p=0.02) and triple-negative cancer (p=0.04).

Conclusions: SUVs for DbPET were equal or superiorto WBPET in predicting the pathological malignancy grade and neoplastic progress in tumors. Furthermore, the presence of RU on DbPET can provide excellent predictive value for high-grade malignancy and might help to determine appropriate therapeutic strategies.

Citation Format: Masumoto N, Kadoya T, Nishina M, Kimura Y, Suzuki E, Sueoka S, Goda N, Sasada S, Kajitani K, Emi A, Haruta R, Kataoka T, Okada M. Evaluation of pathological malignancy grade and neoplastic progress of breast cancer using dedicated breast positron emission tomography [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P5-02-03.

Abstract P6-03-08: Detection ability of dedicated breast positron emission tomography for small-sized breast cancer

Background: Whole body (WB) 18F-fluorodeoxyglucose positron emission tomography (PET) has a relatively poor spatial resolution (&gt;1 cm), which limits the capability to detect small lesions. Therefore, small-sized breast cancers (≤1 cm) may not be visible on WBPET. To overcome these limitations, dedicated breast PET (DbPET) has been developed to improve spatial resolution. DbPET enables detailed high-resolution images within the breast. We aimed to determine whether DbPET can detect small-sized breast cancer compared to WBPET.

Methods: We investigated 203 consecutive patients (217 tumors) (T1–3, N0–3a, M0) with breast cancer who underwent concurrent DbPET and WBPET between January 2016 and March 2017. All DbPET and WBPET images were semi-quantified based on standard uptake values. The diagnostic performance of each scanner was assessed in DbPET and WBPET. Tumors were classified based on pathological classification as follows: Tis, ductal carcinoma in situ (DCIS); T1a, ≤0.5 cm; T1b, 0.5–1 cm; and T1c, 1–2 cm; T2, 2–5 cm; T3, &gt;5 cm. The sensitivities of DbPET and WBPET were compared in each size group.

Results: Table 1 shows the detection rate of breast cancer in WBPET and DbPET

The detection rate of breast cancer in WB- and Db-PET DbPETWBPETpTumor sizeDetection (-) n(%)Detection (+) n(%)Detection (-) n(%)Detection (+) n(%) Tis6(14.6)35(85.4)18(43.9)23(56.1)0.0030T1a2(8)23(92)7(28)18(72)0.0594T1b2(6.5)29(93.5)10(32.3)21(67.7)0.0077T1c5(8.2)56(91.8)11(18)50(82)0.1038T20(0)57(100)1(1.8)56(98.2)0.2375T30(0)2(100)0(0)2(100)-total15(6.9)202(93.1)47(21.7)170(78.3)&lt;0.0001

. The overall detection rate in DBPET [93.1% (202/217)] was significantly higher than that of WBPET [78.3% (170/217)] (P &lt; 0.001). For smaller tumors, DbPET was more sensitive than WBPET: Tis (85.4% vs. 56.1%), T1a (92% vs. 72%), T1b (93.5% vs. 67.7%), T1c (91.8% vs. 82%), T2 (100% vs. 98.2%), and T3 (100% vs. 100%). The sensitivity of DbPET was significantly higher than that of WBPET in Tis (P = 0.003) and T1b (P = 0.008) and tended to be higher than that of WBPET in T1a (P = 0.059). Conversely, no significant differences were observed in T1c (P = 0.103) and T2 (P = 0.238).

Conclusion: The imaging sensitivity of DbPET was higher than that of WBPET. DbPET showed significant sensitivity in DCIS and tumors ≤1 cm, which is a weak point for WBPET. DbPET may serve as a new diagnostic modality to detect small-sized breast cancer.

Citation Format: Sueoka S, Masumoto N, Nishina M, Kimura Y, Suzuki E, Goda N, Sasada S, Kajitani K, Emi A, Haruta R, Kadoya T, Kataoka T, Okada M. Detection ability of dedicated breast positron emission tomography for small-sized breast cancer [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P6-03-08.

Ecological dissimilarity among land-use/land-cover types improves a heterogeneity index for predicting biodiversity in agricultural landscapes

Land-use/land-cover heterogeneity is among the most important factors influencing biodiversity in agricultural landscapes and is the key to the conservation of multi-habitat dwellers that use both terrestrial and aquatic habitats. Heterogeneity indices based on land-use/land-cover maps typically do not integrate ecological dissimilarity between land-use/land-cover types. Here, we applied the concept of functional diversity to an existing land-use/land-cover diversity index (Satoyama index) to incorporate ecological dissimilarity and proposed a new index called the dissimilarity-based Satoyama index (DSI). Using Japan as a case study, we calculated the DSI for three land-use/land-cover maps with different spatial resolutions and derived similarity information from normalized difference vegetation index values. The DSI showed better performance in the prediction of Japanese damselfly species richness than that of the existing index, and a higher correlation between the index and species richness was obtained for higher resolution maps. Thus, our approach to improve the land-use/land-cover diversity index holds promise for future development and can be effective for conservation and monitoring efforts.

Abstract P4-02-06: Evaluation of contrast-enhanced ultrasonography for early prediction of response to neoadjuvant chemotherapy in triple negative breast cancer

Objectives

We aimed to determine whether contrast-enhanced ultrasonography (CEUS) can predict the early effects of neoadjuvant chemotherapy on triple negative breast cancer.

Methods

The clinical responses of 20 consecutive patients with breast cancer (T1–2, N0–1, M0) to neoadjuvant chemotherapy between October 2012 and Feb 2016 were assessed using ultrasonography and contrast-enhanced ultrasonography before starting the therapy and after the treatment of 2 courses. Ascending slope (AS) of perfusion parameters for contrast-enhanced ultrasonography were created from time–intensity curves based on enhancement intensity and temporal changes to objectively evaluate contrast-enhanced ultrasonography findings.We investigated whether rate of change of ascending slope (ΔAS) and tumor size (Δ US) could predict pCR.

Results

Eight (40.0%) of the 20 patients achieved pathological complete response. ΔAS were significantly higher (-25.5 ± 35.5 vs. 14.7 ± 33.2; P &lt; 0.02) in patients who achieved pCR than in those who did not. On the other hand, ΔUS of pCR and non-pCR did not significantly differ among tumors (-40.8 ± 22.4 vs. -21.4 ± 20.6; P = 0.06). The AUC values for ΔAS and ΔUS were 0.792 (95% CI, 0.579 -1.000, P = 0.03) and 0.729 (95% CI, 0.501 - 0.957;p = 0.09), respectively. We set ΔAS and Δ US cut-offs for predicting pCR at-20.08 and - 33.75 based on the ROC curves. Clinical and pathological characteristics of the 20 patients are summarized in Univariate (odds ratio, 2.71; p= 0.02) and multivariate (odds ratio, 2.88; p= 0.03) analysis showed that ΔAS was the sole independent predictor of pCR.

Clinical and pathological characteristics of patients with breast cancer.CharacteristicpCR (n)Non pCR (n)pClinical T status   T1110.71T2711 Clinical N status   Negative570.85Positive35 Nuclear Grade   1 or 2340.85358 ΔUS (%)   &lt; -33.75390.09≥ -33.7553 ΔAS (%)   &lt; -20.082100.009≥ -20.0862 

Univariate and multivariate logistic analysis of significant predictive factors for pCR in triple negative subtype Univariate analysisMultivariate analysis OR, 95% CI, pOR, 95% CI, pΔUS  &gt; -33.751.61, 0.72-34.7, 0.101.85, 0.51-79.1, 0.15≤-33.75  ΔAS  &gt; -20.082.71, 1.65-136.1, 0.022.88, 1.44-218.7, 0.03≤ -20.08  

Conclusion

ΔAS assessed with CEUS can help the physician to early predict the probability of achieving pCR or not.

Citation Format: Masumoto N, Kadoya T, Murakami C, Gouda N, Sasada S, Emi A, Haruta R, Kataoka T, Okada M. Evaluation of contrast-enhanced ultrasonography for early prediction of response to neoadjuvant chemotherapy in triple negative breast cancer [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P4-02-06.