Ryusei Matsuyama

Predicting 5-fluorouracil chemosensitivity of liver metastases from colorectal cancer using primary tumor specimens: Three-gene expression model predicts clinical response

We identified genes related to 5‐fluorouracil (5‐FU) sensitivity in colorectal cancer and utilized these genes for predicting the 5‐FU sensitivity of liver metastases. Eighty‐one candidate genes involved in 5‐FU resistance in gastric and colon cancer cell lines were previously identified using a cDNA microarray. In this study, the mRNA expression levels of these 81 selected genes and the genes of 5‐FU‐related enzymes, including thymidylate synthase (TS), dihydropyrimidine dehydrogenase (DPD) and orotate phosphoribosyltransferase (OPRT), were measured using real‐time quantitative RT‐PCR assays of surgically resected materials from primary colorectal tumors in 22 patients. Clinical responses were estimated by evaluating the effects of 5‐FU‐based hepatic artery injection (HAI) chemotherapy for synchronous liver metastases. Four genes (TNFRSF1B, SLC35F5, NAG‐1 and OPRT) had significantly different expression profiles in 5‐FU‐nonresponding and responding tumors (p < 0.05). A “Response Index” system using three genes (TNFRSF1B, SLC35F5 and OPRT) was then developed using a discriminate analysis; the results were well correlated with the individual chemosensitivities. Among the 11 cases with positive scores in our response index, 9 achieved a reduction in their liver metastases after 5‐FU‐based chemotherapy, whereas only 1 of the 11 cases with negative scores responded well to chemotherapy. Our “Response Index” system, consisting of TNFRSF1B, SLC35F5 and OPRT, has great potential for predicting the efficacy of 5‐FU‐based chemotherapy against liver metastases from colorectal cancer.

Tumor-targeting Salmonella typhimurium A1-R in combination with doxorubicin eradicate soft tissue sarcoma in a patient-derived orthotopic xenograft (PDOX) model

A patient with high grade undifferentiated pleomorphic soft-tissue sarcoma from a striated muscle was grown orthotopically in the right biceps femoris muscle of mice to establish a patient-derived orthotopic xenograft (PDOX) model. Twenty PDOX mice were divided into 4 groups: G1, control without treatment; G2, Salmonella typhimurium (S. typhimurium)A1-R administered by intratumoral (i.t.) injection once a week for 4 weeks; G3, doxorubicin (DOX) administered by intraperitoneal (i.p.) injection once a week for 4 weeks; G4, S. typhimurium A1-R (i.t.) administered once a week for 2 weeks followed by i.p. doxorubicin once a week for 2 weeks. On day 25 from the initiation of treatment, tumor volume in G2, G3, and G4 was significantly lower than G1. Mice found without gross tumor included one mouse (20%) in G2; one mouse (20%) in G3; and 3 mice (60%) in G4. Body weight loss did not significantly differ between the 3 treated groups or from the untreated control. Histological examination revealed eradication of tumor only in G4 where mice were treated with S. typhimurium A1-R followed by DOX. Our present study indicates future clinical potential of combining S. typhimurium A1-R with chemotherapy such as DOX for soft tissue sarcoma patients.

Metastatic recurrence in a pancreatic cancer patient derived orthotopic xenograft (PDOX) nude mouse model is inhibited by neoadjuvant chemotherapy in combination with fluorescence-guided surgery with an anti-CA 19-9-conjugated fluorophore

The aim of this study is to determine the efficacy of neoadjuvant chemotherapy (NAC) with gemcitabine (GEM) in combination with fluorescence-guided surgery (FGS) on a pancreatic cancer patient derived orthotopic xenograft (PDOX) model. A PDOX model was established from a CA19-9-positive, CEA-negative tumor from a patient who had undergone a pancreaticoduodenectomy for pancreatic adenocarcinoma. Mice were randomized to 4 groups: bright light surgery (BLS) only; BLS+NAC; FGS only; and FGS+NAC. An anti-CA19-9 or anti-CEA antibody conjugated to DyLight 650 was administered intravenously via the tail vein of mice with the pancreatic cancer PDOX 24 hours before surgery. The PDOX was brightly labeled with fluorophore-conjugated anti-CA19-9, but not with a fluorophore-conjugated anti-CEA antibody. FGS was performed using the fluorophore-conjugated anti-CA19-9 antibody. FGS had no benefit over BLS to prevent metastatic recurrence. NAC in combination with BLS did not convey an advantage over BLS to prevent metastatic recurrence. However, FGS+NAC significantly reduced the metastatic recurrence frequency to one of 8 mice, compared to FGS only after which metastasis recurred in 6 out of 8 mice, and BLS+NAC with metastatic recurrence in 7 out of 8 mice (p = 0.041). Thus NAC in combination with FGS can reduce or even eliminate metastatic recurrence of pancreatic cancer sensitive to NAC. The present study further emphasizes the power of the PDOX model which enables metastasis to occur and thereby identify the efficacy of NAC in combination with FGS on metastatic recurrence.

Effective molecular targeting of CDK4/6 and IGF-1R in a rare FUS-ERG fusion CDKN2A -deletion doxorubicin-resistant Ewing's sarcoma patient-derived orthotopic xenograft (PDOX) nude-mouse model

Ewings sarcoma is a rare and aggressive malignancy. In the present study, tumor from a patient with a Ewings sarcoma with cyclin-dependent kinase inhibitor 2A/B (CDKN2A/B) loss and FUS-ERG fusion was implanted in the right chest wall of nude mice to establish a patient-derived orthotopic xenograft (PDOX) model. The aim of the present study was to determine efficacy of cyclin-dependent kinase 4/6 (CDK4/6) and insulin-like growth factor-1 receptor (IGF-1R) inhibitors on the Ewings sarcoma PDOX. The PDOX models were randomized into the following groups when tumor volume reached 50 mm3: G1, untreated control; G2, doxorubicin (DOX) (intraperitoneal (i.p.) injection, weekly, for 2 weeks); G3, CDK4/6 inhibitor (palbociclib, PD0332991, per oral (p.o.), daily, for 14 days); G4, IGF-1R inhibitor (linsitinib, OSI-906, p.o., daily, for 14 days). Tumor growth was significantly suppressed both in G3 (palbociclib) and in G4 (linsitinib) compared to G1 (untreated control) at all measured time points. In contrast, DOX did not inhibit tumor growth at any time point, which is consistent with the failure of DOX to control tumor growth in the patient. The results of the present study demonstrate the power of the PDOX model to identify effective targeted molecular therapy of a recalcitrant DOX-resistant Ewings sarcoma with specific genetic alterations. The results of this study suggest the potential of PDOX models for individually-tailored, effective targeted therapy for recalcitrant cancer.

Tumor-Targeting Salmonella typhimurium A1-R Arrests a Chemo-Resistant Patient Soft-Tissue Sarcoma in Nude Mice

A patient-derived nude-mouse model of soft-tissue sarcoma has been established and treated in the following groups: (1) untreated controls; (2) gemcitabine (GEM) (80 mg/kg, ip, weekly, 3 weeks); (3) Pazopanib (100 mg/kg, orally, daily, 3 weeks) and (4) Salmonella typhimurium A1-R (5 × 107 CFU/body, ip, weekly, 3 weeks). The sarcoma was resistant to GEM (p = 0.879). Pazopanib tended to reduce the tumor volume compared to the untreated mice, but there was no significant difference (p = 0.115). S. typhimurium A1-R significantly inhibited tumor growth compared to the untreated mice (p = 0.001). S. typhimurium A1-R was the only effective treatment for the soft-tissue sarcoma nude mouse model among all treatments including a newly approved multiple tyrosine kinase inhibitor; Pazopanib. These results suggest tumor-targeting S. typhimurium A1-R is a promising treatment for chemo-resistant soft-tissue sarcoma.

Influence of Chemotherapy on Liver Regeneration Induced by Portal Vein Embolization or First Hepatectomy of a Staged Procedure for Colorectal Liver Metastases

BackgroundAlthough portal vein embolization (PVE) and staged hepatectomy (StHx), as well as prehepatectomy chemotherapy, have improved the resectability rate of patients with multiple bilobar colorectal liver metastases, the impact of prehepatectomy chemotherapy on liver hypertrophy following PVE and/or StHx has remained unclear.MethodsSixty patients who underwent PVE followed by one-stage hepatectomy and StHx with or without PVE were analyzed. Liver hypertrophy following PVE and/or the first hepatectomy of StHx and the clinical course after final hepatectomy was compared between patients with and without prehepatectomy chemotherapy.ResultsNo difference of volume of the future liver remnant (FLR) before or after the procedure was seen between the chemotherapy group and the nonchemotherapy group. Even in 38 patients who underwent right PVE prior to a planned right hemihepatectomy, the chemotherapy group (n = 14) and the nonchemotherapy group (n = 24) were comparable in terms of volumes of FLR before (P = 0.71) and after (P = 0.29) PVE and posthepatectomy courses. However, the liver hypertrophy ratio for patients showing steatosis in adjacent nonmalignant liver parenchyma, which frequently is induced by chemotherapy, was lower than that for patients without steatosis (P = 0.04).ConclusionsAlthough prehepatectomy chemotherapy did not impair liver hypertrophy, PVE and/or StHx accompanied by prehepatectomy chemotherapy should be performed with particular care to minimize risk of liver failure after the procedure.

Tumor-targeting Salmonella typhimurium A1-R regresses an osteosarcoma in a patient-derived xenograft model resistant to a molecular-targeting drug

Osteosarcoma occurs mostly in children and young adults, who are treated with multiple agents in combination with limb-salvage surgery. However, the overall 5-year survival rate for patients with recurrent or metastatic osteosarcoma is 20-30% which has not improved significantly over 30 years. Refractory patients would benefit from precise individualized therapy. We report here that a patient-derived osteosarcoma growing in a subcutaneous nude-mouse model was regressed by tumor-targeting Salmonella typhimurium A1-R (S. typhimurium A1-R, p<0.001 compared to untreated control). The osteosarcoma was only partially sensitive to the molecular-targeting drug sorafenib, which did not arrest its growth. S. typhimurium A1-R was significantly more effective than sorafenib (P <0.001). S. typhimurium grew in the treated tumors and caused extensive necrosis of the tumor tissue. These data show that S. typhimurium A1-R is powerful therapy for an osteosarcoma patient-derived xenograft model.

Tumor-Targeting Salmonella typhimurium A1-R in Combination with Trastuzumab Eradicates HER-2-Positive Cervical Cancer Cells in Patient-Derived Mouse Models

We have previously developed mouse models of HER-2-positive cervical cancer. Tumors in nude mice had histological structures similar to the original tumor and were stained by anti-HER-2 antibody in the same pattern as the patient’s cancer. We have also previously developed tumor-targeting Salmonella typhimurium A1-R and have demonstrated its efficacy against patient-derived tumor mouse models, both alone and in combination. In the current study, we determined the efficacy of S. typhimurium A1-R in combination with trastuzumab on a patient-cancer nude-mouse model of HER-2 positive cervical cancer. Mice were randomized to 5 groups and treated as follows: (1) no treatment; (2) carboplatinum (30 mg/kg, ip, weekly, 5 weeks); (3) trastuzumab (20 mg/kg, ip, weekly, 5 weeks); (4) S. typhimurium A1-R (5 × 107 CFU/body, ip, weekly, 5 weeks); (5) S. typhimurium A1-R (5 × 107 CFU/body, ip, weekly, 5 weeks) + trastuzumab (20 mg/kg, ip, weekly, 5 weeks). All regimens had significant efficacy compared to the untreated mice. The relative tumor volume of S. typhimurium A1-R + trastuzumab-treated mice was smaller compared to trastuzumab alone (p = 0.007) and S. typhimurium A1-R alone (p = 0.039). No significant body weight loss was found compared to the no treatment group except for carboplatinum-treated mice (p = 0.021). Upon histological examination, viable tumor cells were not detected, and replaced by stromal cells in the tumors treated with S. typhimurium A1-R + trastuzumab. The results of the present study suggest that S. typhimurium A1-R and trastuzumab in combination are highly effective against HER-2-expressing cervical cancer.

Fluorescence-guided surgery, but not bright-light surgery, prevents local recurrence in a pancreatic cancer patient derived orthotopic xenograft (PDOX) model resistant to neoadjuvant chemotherapy (NAC)

BACKGROUND The aim of this study is to determine the efficacy of neoadjuvant chemotherapy (NAC) with gemcitabine (GEM) in combination with fluorescence-guided surgery (FGS) on a pancreatic cancer patient derived orthotopic xenograft (PDOX) model. METHODS A PDOX model was established from a CEA-positive tumor from a patient who had undergone a pancreaticoduodenectomy for pancreatic adenocarcinoma. Mice were randomized to 4 groups: bright light surgery (BLS) only; BLS + NAC; FGS only; and FGS + NAC. An anti-CEA antibody conjugated to DyLight 650 was administered intravenously via the tail vein of mice with a pancreatic cancer PDOX 24 h before surgery. RESULTS The PDOX was clearly labeled with fluorophore-conjugated anti-CEA antibody. Only one out of 8 mice had local recurrence in the FGS only group and zero out of 8 mice had local recurrence in the FGS + NAC which was significantly lower than BLS only or BLS + NAC mice, where local disease recurred in 6 out of 8 mice in each treatment group (p = 0.041 and p = 0.007, respectively). NAC did not significantly reduce recurrence rates when combined with either FGS or BLS. CONCLUSION These results indicate that FGS can significantly reduce local recurrence compared to BLS in pancreatic cancer resistant to NAC.

Imaging and surgical planning for perihilar cholangiocarcinoma

Recent advances in multidetector computed tomography (MDCT) offer several benefits for management of perihilar tumors. Resection planning for perihilar cholangiocarcinoma should consider two factors: safety and curability. Recognition of individual anatomic variations is particularly important for avoiding intraoperative injury. In particular, hepatic arterial variations often restrict resection procedures. Extent of both longitudinal and vertical invasion by biliary tumors can be estimated from multiplanar reconstruction (MPR) images. Longitudinal extent of resection can be planned based on two anatomic landmarks, the U point and the P point, readily identifiable in preoperative 3‐dimensional (3D) images and by intraoperative inspection. Concerning vertical invasion, when direct vascular invasion is suspected from a finding of attachment of tumor and vessels such as portal veins and/or hepatic arteries without a thin low‐density plane of separation shown by MPR, these vessels should be resected en bloc with the tumor. Surgical team members can plan and simulate details of vascular resection and reconstruction using 3D images. Reduced operative morbidity and increased R0 resection rates are expected because of better planning of procedures. These techniques soon may increase long‐term survival for patients with perihilar cholangiocarcinoma.