Stromal disrupting effects of nab-paclitaxel in pancreatic cancer

Background:

Nab-paclitaxel and gemcitabine have demonstrated a survival benefit over gemcitabine alone in advanced pancreatic cancer (PDA). This study aimed to investigate the clinical, biological, and imaging effects of the regimen in patients with operable PDA.

Methods:

Patients with operable PDA received two cycles of nab-paclitaxel and gemcitabine before surgical resection. FDG-PET and CA19.9 tumour marker levels were used to measure clinical activity. Effects on tumour stroma were determined by endoscopic ultrasound (EUS) elastography. The collagen content and architecture as well as density of cancer-associated fibroblasts (CAFs) were determined in the resected surgical specimen and compared with a group of untreated and treated with conventional chemoradiation therapy controls. A co-clinical study in a mouse model of PDA was conducted to differentiate between the effects of nab-paclitaxel and gemcitabine.

Results:

A total of 16 patients were enrolled. Treatment resulted in significant antitumour effects with 50% of patients achieving a >75% decrease in circulating CA19.9 tumour marker and a response by FDG-PET. There was also a significant decrement in tumour stiffness as measured by EUS elastography. Seven of 12 patients who completed treatment and were operated had major pathological regressions. Analysis of residual tumours showed a marked disorganised collagen with a very low density of CAF, which was not observed in the untreated or conventionally treated control groups. The preclinical co-clinical study showed that these effects were specific of nab-paclitaxel and not gemcitabine.

Conclusion:

These data suggest that nab-paclitaxel and gemcitabine decreases CAF content inducing a marked alteration in cancer stroma that results in tumour softening. This regimen should be studied in patients with operable PDA.     

Effects of a Ketogenic Diet on [18F]FDG-PET Imaging in a Mouse Model of Lung Cancer

Purpose

Myocardial uptake can hamper visualization of lung tumors, atherosclerotic plaques, and inflammatory diseases in 2-deoxy-2-[¹⁸F]fluoro-D-glucose ([¹⁸F]FDG) studies because it leads to spillover in adjacent structures. Several preparatory pre-imaging protocols (including dietary restrictions and drugs) have been proposed to decrease physiological [¹⁸F]FDG uptake by the heart, although their effect on tumor glucose metabolism remains largely unknown. The objective of this study was to assess the effects of a ketogenic diet (as an alternative protocol to fasting) on tumor glucose metabolism assessed by [¹⁸F]FDG positron emission tomography (PET) in a mouse model of lung cancer.

Procedures

PET scans were performed 60 min after injection of 18.5 MBq of [¹⁸F]FDG. PET data were collected for 45 min, and an x-ray computed tomograph (CT) image was acquired after the PET scan. A PET/CT study was obtained for each mouse after fasting and after the ketogenic diet. Quantitative data were obtained from regions of interest in the left ventricular myocardium and lung tumor.

Results

Three days on a ketogenic diet decreased mean standard uptake value (SUVmean) in the myocardium (SUVmean 0.95?±?0.36) more than one night of fasting (SUVmean 1.64?±?0.93). Tumor uptake did not change under either dietary condition.

Conclusions

These results show that 3 days on high-fat diets prior to [¹⁸F]FDG-PET imaging does not change tumor glucose metabolism compared with one night of fasting, although high-fat diets suppress myocardial [¹⁸F]FDG uptake better than fasting.

     

Complete Regression of Advanced Pancreatic Ductal Adenocarcinomas upon Combined Inhibition of EGFR and C-RAF

1. Download high-res image (248KB)
2. Download full-size image

     

KRAS4A induces metastatic lung adenocarcinomas in vivo in the absence of the KRAS4B isoform

In mammals, the KRAS locus encodes two protein isoforms, KRAS4A and KRAS4B, which differ only in their C terminus via alternative splicing of distinct fourth exons. Previous studies have shown that whereas KRAS expression is essential for mouse development, the KRAS4A isoform is expendable. Here, we have generated a mouse strain that carries a terminator codon in exon 4B that leads to the expression of an unstable KRAS4B¹⁵⁴ truncated polypeptide, hence resulting in a bona fide Kras4B-null allele. In contrast, this terminator codon leaves expression of the KRAS4A isoform unaffected. Mice selectively lacking KRAS4B expression developed to term but died perinatally because of hypertrabeculation of the ventricular wall, a defect reminiscent of that observed in embryos lacking the Kras locus. Mouse embryonic fibroblasts (MEFs) obtained from Kras4B^−/− embryos proliferated less than did wild-type MEFs, because of limited expression of KRAS4A, a defect that can be compensated for by ectopic expression of this isoform. Introduction of the same terminator codon into a Kras^FSFG12V allele allowed expression of an endogenous KRAS4A^G12V oncogenic isoform in the absence of KRAS4B. Exposure of Kras^{+/FSF4AG12V4B–} mice to Adeno-FLPo particles induced lung tumors with complete penetrance, albeit with increased latencies as compared with control Kras^+/FSFG12V animals. Moreover, a significant percentage of these mice developed proximal metastasis, a feature seldom observed in mice expressing both mutant isoforms. These results illustrate that expression of the KRAS4A^G12V mutant isoform is sufficient to induce lung tumors, thus suggesting that selective targeting of the KRAS4B^G12V oncoprotein may not have significant therapeutic consequences.

The KRAS locus has gained prominence due to its involvement in as many as one-fourth of all human tumors (1, 2). This locus encodes two protein isoforms, KRAS4A and KRAS4B, via alternative splicing of independent fourth exons. Both proteins share almost identical amino acids from residues 1 to 164, a region that includes the guanosine triphosphate–binding domain, as well as the switch I and II regions responsible for engaging with their main downstream effectors (3–5). These residues also include all known oncogenic mutations. However, these isoforms differ substantially from residue 165 onward, a region that includes the hypervariable region (HVR) primarily involved in protein trafficking as well as in establishing its location within the plasma membrane (5, 6). Finally, both isoforms are prenylated at the cysteine residue of their respective CAAX boxes located at their C termini.The KRAS4B protein contains a unique polybasic stretch not found in other RAS proteins formed by a series of lysine residues that presumably organizes its localization to distinct nanoclusters. In addition, KRAS4B can be phosphorylated at serine 181 within the polybasic stretch, which is believed to reduce the net charge of the HVR, thus decreasing its affinity for the plasma membrane (4). Binding of calmodulin to KRAS4B is known to block this phosphorylation step, although the functional significance remains a matter of debate (7, 8). Interestingly, calmodulin binding to oncogenic KRAS4B suppressed noncanonical Wnt signaling. Disruption of this interaction has been proposed to reduce the tumorigenic properties of mutant KRAS4B in pancreatic cancer (9). KRAS4A on the other hand uses a hybrid membrane-targeting motif consisting of an acylatable cysteine residue as well as a bipartite polybasic region and resides in disorganized liquid domains (10, 11). Moreover, lysine fatty acylation selectively regulates membrane distribution and in vitro transforming properties of KRAS4A (12). The distinct membrane distribution properties of KRAS4A have also been attributed to its association with HK1 in the outer mitochondrial membrane, an activity that is not shared by KRAS4B (13).Despite these differences in their HVR, the biological significance of the two KRAS isoforms is less clear. Whereas genetic disruption of the entire Kras locus causes embryonic lethality at midgestation, selective ablation of the KRAS4A isoform did not result in any obvious phenotype either during embryonic development or adult homeostasis, thus suggesting that expression of KRAS4B is sufficient to perform most of the KRAS functions in mice (14–16). Indeed, KRAS4B appears to represent the majority of the two isoforms during embryonic development, while KRAS4A is more dynamically regulated and its ratio slightly increases in several adult tissues (17). Several studies, however, have detected widespread expression of KRAS4A in RAS mutant cancers (10, 18, 19). In addition, absence of KRAS4A reduced tumorigenesis when using a chemical carcinogenesis model of lung cancer (20). Finally, knockin of a KRAS4B complementary DNA (cDNA) into the Kras locus following a strategy that eliminates KRAS4A expression rendered mice largely resistant to chemical carcinogenesis (21).To better understand the unique role(s) of the KRAS4A isoform in normal and tumor-bearing mice, we have generated wild-type and oncogenic Kras alleles that prevent expression of the KRAS4B isoform without altering the overall genomic structure of the Kras locus, hence not affecting expression of either wild-type or mutant KRAS4A isoforms, respectively. These studies provide important insights into the role of the KRAS4A isoform both during embryonic development and in tumor formation.