C. Charnsangavej

Correlation of Computed Tomography and Positron Emission Tomography in Patients With Metastatic Gastrointestinal Stromal Tumor Treated at a Single Institution With Imatinib Mesylate: Proposal of New Computed Tomography Response Criteria

PURPOSE Response Evaluation Criteria in Solid Tumors (RECIST) are insensitive in evaluating gastrointestinal stromal tumors (GISTs) treated with imatinib. This study evaluates whether computed tomography (CT) findings of GIST after imatinib treatment correlate with tumor responses by [18F]fluorodeoxyglucose (FDG) positron emission tomography (PET) and develops reliable, quantitative, CT response criteria. PATIENTS AND METHODS A total of 172 lesions selected by RECIST were evaluated in 40 patients with metastatic GISTs treated with imatinib. All patients had pretreatment and 2-month follow-up CTs and FDG-PETs. Multivariate analysis was performed using tumor size and density (Hounsfield unit [HU]) on CT and maximum standardized uptake value (SUVmax) on FDG-PET. Patients were observed up to 28 months. RESULTS Mean baseline tumor size and density on CT were 5.3 cm and 72.8 HU, respectively, and mean baseline SUVmax on FDG-PET was 5.8. Thirty-three patients had good response on FDG-PET. A decrease in tumor size of more than 10% or a decrease in tumor density of more than 15% on CT had a sensitivity of 97% and a specificity of 100% in identifying PET responders versus 52% and 100% by RECIST. Good responders on CT at 2 months had significantly longer time to progression than those who did not respond (P = .01). CONCLUSION Small changes in tumor size or density on CT are sensitive and specific methods of assessing the response of GISTs. If the prognostic value of our proposed CT response criteria can be confirmed prospectively, the criteria should be employed in future studies of patients with GIST.

We Should Desist Using RECIST, at Least in GIST

PURPOSE Response Evaluation Criteria in Solid Tumors (RECIST) are insensitive in evaluating imatinib-treated gastrointestinal stromal tumors (GISTs). Response by Choi criteria, a 10% decrease in size or a 15% decrease in density on contrast-enhanced CT, correlated well in a small training set of patients who showed response as measured by positron emission tomography, and was more predictive of time to tumor progression (TTP) than response by RECIST. This study was designed to validate these observations in an independent data set. PATIENTS AND METHODS Fifty-eight patients with imatinib-treated GISTs were evaluated by RECIST and Choi criteria. TTP was compared with TTP in the training set. Patients were analyzed initially with follow-up to 28 months, extended to 60 months for survival analysis. RESULTS Patients who met Choi response criteria on CT at 2 months had significantly better TTP than those who did not (P = .0002), whereas response group by RECIST was not significantly correlated with TTP. Even when the 98 patients from both sets were analyzed together, the response group by RECIST did not correlate significantly with TTP, whereas response group by Choi criteria did correlate significantly with TTP. Disease-specific survival (DSS) was also significantly correlated with response group by Choi criteria (P = .04), but not with response group by RECIST. CONCLUSION Choi response criteria are reproducible, more sensitive, and more precise than RECIST in assessing the response of GISTs to imatinib mesylate. Response by Choi criteria, unlike response by RECIST, correlates significantly with TTP and DSS. Response by Choi criteria should be incorporated routinely into future studies of GIST therapy. We should desist using RECIST, at least in GIST.

Is the therapeutic index better with gemcitabine-based chemoradiation than with 5-fluorouracil-based chemoradiation in locally advanced pancreatic cancer?

PURPOSE To retrospectively compare the toxicity and efficacy of concurrent gemcitabine-based chemoradiation with that of concurrent 5-fluorouracil (5-FU)-based chemoradiation in patients with unresectable pancreatic cancer. PATIENTS AND METHODS Between September 1996 and May 2000, 114 patients with localized unresectable adenocarcinoma of the pancreas were treated with concurrent chemoradiation. Locally advanced unresectable disease was defined as low-density tumor in contact with the superior mesenteric artery (SMA) or celiac artery, or occlusion of the superior mesenteric-portal venous confluence. Fifty-three patients were selected to receive gemcitabine in 7 weekly cycles (250-500 mg/m(2)) with concurrent radiotherapy (median dose 30 Gy, range 30-33 Gy in 10-11 fractions). The remaining 61 patients received continuous-infusion 5-FU (200-300 mg/m(2)) with concurrent radiotherapy (30 Gy in 10 fractions). Radiotherapy was delivered to the primary tumor and regional lymphatics. Patients receiving gemcitabine and those receiving 5-FU had a similar mean Karnofsky performance status (KPS, 89% vs. 86%), distribution of tumor grade (43% vs. 33% poorly differentiated), and percent weight loss (all p = NS). However, patients treated with gemcitabine had a significantly larger median maximum cross-sectional tumor area (TA, 8.8 cm(2) vs. 5.7 cm(2), p = 0.046) and were significantly younger (median age 60 vs. 68 years, p <0.001). Severe acute toxicity (ST) was defined as toxicity requiring a hospital stay of more than 5 days, mucosal ulceration with bleeding, more than 3 dose deletions of gemcitabine or discontinuation of 5-FU, or toxicity resulting in surgical intervention or death. Kaplan-Meier analysis was used to calculate the actuarial rate of local progression on imaging (LP), the rate of distant metastasis (DM), and the overall survival (OS) rate. The imaging was reviewed in resected patients. RESULTS Patients receiving gemcitabine developed significantly more ST during treatment (23% vs. 2%, p < 0.0001) than did those receiving 5-FU. Patients treated with gemcitabine had a similar 10-month LP rate (62% vs. 61%), 10-month DM rate (55% vs. 47%), 1-year OS rate (42% vs. 28%), and median OS duration (11 months vs. 9 months) to patients treated with 5 FU (all p = NS). Five patients who received gemcitabine and 1 patient who received 5-FU underwent margin-negative pancreaticoduodenectomy after chemoradiation. Three patients had a short segment (<or= 1 cm in length) of low-density tumor abutting the SMA, 1 had involvement of the common hepatic artery, and 1 had a short-segment occlusion of the superior mesenteric vein, amenable to venous resection and reconstruction. The other patient was thought to have inflammatory changes discontiguous with the tumor surrounding the SMA, which resolved after therapy. TA >10 cm(2) (p = 0.03) and poor differentiation (p = 0.07) were associated with a worse survival duration; however, other factors, such as KPS and weight loss >10% and age did not influence OS. CONCLUSION Despite the selection of healthier patients to receive gemcitabine, there was a significantly higher severe toxicity rate than with 5-FU. The median and 1-year survivals were not significantly different with the use of concurrent gemcitabine; however, the tumors treated were significantly larger. Additionally, a small number of patients with minimal arterial involvement whose disease met our radiographic definition of unresectable disease had margin-negative resections after treatment with gemcitabine-based chemoradiation. These possible benefits and the high rate of severe toxicity define a very narrow therapeutic index for concurrent gemcitabine-based chemoradiation given by this schedule of administration.

Vascular involvement in pancreatic adenocarcinoma:reassessment by thin-section CT

Abstract. We defined computed tomographic (CT) criteria of vascular involvement by pancreatic carcinoma and used these criteria to assess vascular involvement in 56 patients with pancreatic adenocarcinoma. CT of the pancreas was performed at 1.5-mm section thickness and 5-mm section intervals during a bolus phase of intravenous contrast enhancement. The type of vascular involvement was correlated with surgical and pathologic findings. When there was fat-plane (type A) or normal pancreatic parenchyma (type B) separating the tumor from adjacent vessels, the tumor could be resected without venous resection in 21 of 22 patients (95%). When the tumor was inseparable from the vessels but the points of contact formed a convexity against the vessel (type C), CT was not reliable in predicting whether or not the tumor was fixed against the vessel. When the tumor was partially encircling (type D) the vessel, the tumor was fixed against the vessels in most cases. The resectable rate was 47%, but resection would also require venous resection. When the tumor was completely encircling (type E) or occluding (type F<+>) the vessel, all tumors were not resectable with a negative margin. Thin-section CT with bolus intravenous contrast enhancement improved the ability to assess vascular involvement in pancreatic adenocarcinoma.

Phase II trial of intravenous flourouracil and subcutaneous interferon alfa-2b for biliary tract cancer.

PURPOSE To assess the efficacy of systemic intravenous-fluorouracil (5-FU) and subcutaneous recombinant human interferon alfa-2b (rIFN alpha-2b) in patients with measurable cancer of the biliary tree. PATIENTS AND METHODS Thirty-five patients (25 with cholangiocarcinoma and 10 with gallbladder carcinoma) were registered onto this phase II protocol between 1992 and 1995. Patients received a continuous infusion of 750 mg/m2/d of 5-FU on days 1 through 5 through a centrally placed venous catheter and a subcutaneous injection of 5 MU/m2 of rIFN alpha-2b on days 1, 3, and 5. Treatment cycles were repeated every 14 days; one course of therapy included four treatment cycles. Disease status was assessed every 8 weeks. Dosages were lowered for grade III mucositis. Fourteen patients had prior treatment and, before initiating this therapy, 17 patients required decompression of the biliary tree. RESULTS Eleven of 32 (34%) assessable patients had a partial response. The median time to disease progression was 9.5 months, and the median survival time 12 months. Grade III to IV toxic effects were granulocytopenia (14%), mucositis (20%), diarrhea (9%), and dermatitis (11%). Grade III to IV asthenia and fatigue were observed in 6% of patients. CONCLUSION Drug tolerance was better among previously untreated patients. To achieve a complete response, additional chemotherapy or radiotherapy should be considered when liver resection or transplantation is not feasible. However, if these results can be reproduced by other investigators, the regimen should be studied for adjuvant treatment of gallbladder carcinoma incidentally identified in patients undergoing cholecystectomy.

Islet Cell Tumors Metastatic to the Liver: Effective Palliation by Sequential Hepatic Artery Embolization

The value of sequential percutaneous hepatic artery embolization with polyvinyl alcohol particles was examined in 22 patients with islet cell carcinoma metastatic to the liver. Nine patients had gastrinoma, 2 had glucagonoma, and 11 had no discernible hormonal secretions or syndromes. Ninety-seven embolizations were done with a median number of 4 (range, 1 to 12) per patient. The interval between embolizations ranged from 1 to 8 months. Twelve of twenty evaluable patients had a partial remission, frequently associated with subjective improvement and decrease in hormone levels. The projected median survival of all 22 patients from the initiation of embolization is 33.7 months (range, 1 to 72). Nausea, vomiting, fever, and abdominal pain occurred with each embolization and subsided usually by day 10 (range, 3 to 35). Sequential hepatic artery occlusion is an effective method for prolonged palliation in this selected group of patients.

Toxicity and Efficacy of Concurrent Gemcitabine and Radiotherapy for Locally Advanced Pancreatic Cancer.

SummaryBackground. Gemcitabine and radiotherapy are a potent combination. A clinical assessment of the therapeutic ratio for locally advanced pancreatic cancer patients has not yet been reported.Aim of Study. To assess the toxicity, survival, and pattern of failure of locally advanced pancreatic cancer patients treated with concurrent gemcitabine-based chemoradiation.Patients and Methods. Between the dates of December 1996 and August 2000 51 patients with locally advanced unresectable adenocarcinoma of the pancreas were treated with concurrent gemcitabine and radiotherapy at MDACC. Patients received 250–500 mg/m2 of gemcitabine weekly ×7 over 30 min and 30–33 Gy in 10–11 fractions over two weeks to the primary tumor and regional lymphatics. Severe toxicity was defined as admission >5 d, mucosal ulceration, >3 dose deletions of gemcitabine or toxicity resulting in surgical intervention or that resulted in death.Results. The median survival was 11 mo. Overall, 37 of 51 patients had objective evidence of local progression. The actuarial rate of local progression rate at 9 mo was 70%. The 9-mo distant metastasis rate was 52%. Tumors ≥ 10 cm2 had worse local control, distant control, and overall survival. Six patients underwent pancreaticoduodenectomy after therapy. After review of the imaging, only four of these patients had minimal arterial involvement, one was incorrectly staged, and one had initial inflammatory change on CT that resolved. Twelve of 51 (24%) patients suffered severe acute toxicity, and 17 of 51 (33%) patients were admitted for supportive care.Conclusion. Concurrent gemcitabine and radiotherapy can be a very difficult combination to administer safely. Our results do not suggest a prolongation of median survival for patients with localized pancreatic cancer treated with this therapy. It is possible that gemcitabine-based chemoradiation contributes to the margin-negative resectability of a small number of patients with minimal arterial involvement, but this benefit is obscured by the frequent toxicity encountered in most patients. Locally advanced pancreatic cancer patients should continue to be enrolled on prospective studies investigating novel combinations of cytotoxic and/or biologic agents with concurrent radiotherapy.

Hepatic Arterial Infusion of Floxuridine, Leucovorin, Doxorubicin, and Cisplatin for Hepatocellular Carcinoma: Effects of Hepatitis B and C Viral Infection on Drug Toxicity and Patient Survival

PURPOSE To conduct a pilot trial of hepatic arterial infusion (HAI) of floxuridine (FUDR), leucovorin, Adriamycin (doxorubicin; Adria Laboratories, Columbus, OH), and cisplatin (FLAP) in nonresectable hepatocellular cancer (HCC) confined to the liver and assess the effects of hepatitis B (HBV) and hepatitis C (HCV) viral markers on toxicity, response to treatment, and patient survival. PATIENTS AND METHODS Of 31 HCC patients, 13 were HBV- and HCV-nonreactive, and 18 had evidence of either current or prior HBV and/or HCV infection. Treatment was delivered through percutaneous hepatic arterial catheters, and Infusaid pumps (Shiley Infusaid, Norwood, MA) were placed in responding patients. Cisplatin (100 mg/m2) and Adriamycin (30 to 35 mg/m2) were administered on day 1, followed by a continuous 24-hour HAI of an admixture of floxuridine (60 mg/m2) and leucovorin (15 mg/m2) daily for 4 days. Treatment was repeated every 5 weeks. RESULTS Twelve (41%) of 29 assessable patients had a partial response (PR), with a median time to disease progression of 13 months. Six (50%) of 12 HBV-negative (HBV-)/HCV-negative (HCV-) and six of 17 (35%) HBV-positive (HCV+) and/or HCV-positive (HCV+) patients achieved a PR. Eight patients have been maintained in remission for a median duration greater than 15.5 months. The median survival duration of all 31 patients was 15 months, 7.5 months among HBV+ and/or HCV+ patients, and significantly longer among hepatitis-non-reactive patients (P = .007). (A median has not yet been reached.) Granulocylopenia (< 0.1 x 10(3)/microL), thrombocytopenia (< 25 x 10(3)/microL), and hospitalizations for infectious complications were significantly more common among HBV-HCV-reactive than -nonreactive patients: 56%, 50%, and 67% versus 15%, 15%, and 8%, respectively (P < .05 for all). CONCLUSION HAI of FLAP has induced long-term PR and has palliated extensive nonresectable HCC. Positive hepatitis serology appeared to increase bone marrow susceptibility to myelotoxic drugs. Conceivably, one or both viruses may have a direct inhibitory effect on bone marrow progenitors and thereby contribute to the observed myelotoxicity.

Multidetector-row helical CT and advanced postprocessing techniques for the evaluation of pancreatic neoplasms

AbstractAn important feature of multidetector-row helical computed tomography (CT) is the increased speed of scanning that permits routine use of very thin collimation and acquisition of near isometric imaging data of the abdomen within the time span of a single breath-hold. The parallel escalation in the capabilities of workstations makes feasible the practical use of advanced postprocessing techniques to create high quality volumetric imaging. This article highlights the unique contributions of multidetector-row CT and advanced postprocessing techniques to the evaluation of the pancreas and peripancreatic vascular structures and their value in the diagnosis and staging of pancreatic neoplasms.

Staging of pancreatic cancer with multidetector CT in the setting of preoperative chemoradiation therapy

BackgroundPreoperative chemoradiation can potentially improve outcomes in patients with pancreatic cancer. This study addresses its effect on staging pancreatic cancer with multidetector computed tomography (MDCT).MethodsFifty-five patients underwent a dual-phase MDCT pancreas protocol for proved pancreatic cancer. Of these, 16 patients underwent preoperative chemoradiation. Three radiologists independently reviewed images to assess for locally advanced disease, liver and peritoneal metastases on baseline studies of all 55 patients, and on follow-up preoperative studies for the 16 patients receiving preoperative therapy. Overall score for resectability was graded on a scale from 1 to 5 (1, definitely resectable; 5. definitely unresectable). Receiver operating characteristic curves and weighted (κ statistics were determined.ResultsThe areas under the receiver operating characteristic curves for readers 1, 2, and 3 were 0.98, 0.96, and 0.90, respectively. Weighted κ values for reader 1 versus reader 2, reader 1 versus reader 3, and reader 2 versus reader 3 were 0.90, 0.57, and 0.54, respectively. Interpreting scores of 1 to 3 for resectability as resectable disease, the mean values for sensitivity, specificity, negative predictive value, positive predictive value, and accuracy were 0.92, 0.91, 0.74, 0.98, and 0.92 respectively.ConclusionThe negative predictive value for MDCT for identifying unresectable pancreatic cancer in the setting of preoperative therapy is comparable to that reported in the absence of neoadjuvant therapy.