- © 2008 by American Society of Clinical Oncology
Phase I Trial of the Human Immunodeficiency Virus Protease Inhibitor Nelfinavir and Chemoradiation for Locally Advanced Pancreatic Cancer
- Thomas B. Brunner,
- Matthias Geiger,
- Gerhard G. Grabenbauer,
- Marga Lang-Welzenbach,
- Tine S. Mantoni,
- Alexander Cavallaro,
- Rolf Sauer,
- Werner Hohenberger and
- W. Gillies McKenna
- From the Departments of Radiation Oncology and Surgery, and Department of Nuclear Medicine, Friedrich-Alexander University of Erlangen-Nuremberg, Nuremberg, Germany; and Gray Institute of Radiation Oncology and Biology, University of Oxford, Oxford, United Kingdom
- Corresponding author: Thomas B. Brunner, MD, Gray Institute of Radiation Oncology and Biology, University of Oxford, Radiobiology Research Institute, Churchill Hospital, Oxford, OX3 7LJ, United Kingdom; e-mail: thomas.brunner{at}rob.ox.ac.uk
Abstract
Purpose Preclinically, HIV protease inhibitors radiosensitize tumors with activated PI3-kinase/Akt pathway. We determined the toxicity of nelfinavir chemoradiotherapy in borderline resectable and unresectable pancreatic cancer.
Patients and Methods Oral nelfinavir (2 × 1,250 mg) was started 3 days before and continued throughout chemoradiotherapy to 50.4 Gy (boost, 59.4 Gy) in 12 patients. Two gemcitabine dose levels (DL) were tested (200 mg/m2 and 300 mg/m2 on days 1, 8, 22, and 29). Cisplatin was administered on the same days at 30 mg/m2. Phospho-Akt downregulation by nelfinavir was monitored by immunoblotting in patient leukocytes. Restaging positron emission tomography (PET)/computed tomography (CT) and CA19-9 levels served to assess response, and responding tumors were resected.
Results At each DL, five of six patients completed chemoradiotherapy, and two of 12 patients had incomplete chemoradiotherapy because of clinical depression (DL1) and peritoneal metastasis (DL2). Grade 4 toxicities were a transaminase elevation (DL2) as a result of biliary stent occlusion and acute cholecystitis as a result of peritoneal metastasis (DL2). Stent occlusions led to dose-limiting toxicities of grade 3 liver enzyme and bilirubin elevations (two patients at DL1, one patient at DL2). Grade 3 nausea and vomiting occurred in a DL2 patient, and weight loss occurred in a DL1 patient who refused supportive feeding. Secondary complete resection was possible in six of 10 patients with complete chemoradiotherapy, including one tumor with pathologic sterilization. Partial CT responses were observed in five of 10 patients who completed chemoradiotherapy. Of nine patients assessable by PET,responses were complete in five patients and partial patients, and stable disease was observed in two patients.
Conclusion The combination of nelfinavir and chemoradiotherapy showed acceptable toxicity and promising activity in patients with pancreatic cancer.
INTRODUCTION
The prognosis of patients with pancreatic ductal adenocarcinoma (PDAC) is poor and has improved only modestly during the last decade through the use of gemcitabine.1 PDAC has the shortest mean European 5-year relative survival rate by cancer site.2 More than 80% of the tumors cannot be resected at presentation, although half of these have no distant metastasis (locally advanced pancreatic adenocarcinoma [LAPC]). Because the disease is still locoregional at that time, chemoradiotherapy can be used to treat LAPC to increase survival. The survival times that can be achieved with this approach at best reach a median of 15 months,3 pointing to the critical need for a new therapeutic approach.
The molecular changes in PDAC have been well characterized, and mutations of K-ras, p53, p16, and DPC4 are key events. K-ras is mutant in up to 90% of all tumors and has been shown to contribute to radiation resistance through autocrine activation of H-ras in preclinical studies.4-6 Other upstream signals like epidermal growth factor receptor and insulin-like growth factor receptor also feed into the PI3-kinase (PI3K)/Akt cascade. Furthermore, amplification of Akt2 has been described in up to 20% of PDAC.7 Overexpression of phospho-Akt was shown to correlate with short survival after surgery.8 No suitable clinical inhibitors of PI3K or Akt have been identified to date. As an example, the widely used PI3K inhibitors LY294002 and wortmannin, which have radiosensitizing properties in vitro, are too toxic to be used in vivo.9 Because this pathway is usually not constitutively activated in normal cells, its inhibition should not result in increased normal tissue toxicity. In addition, inhibition of this pathway has been shown to enhance gemcitabine-induced (Gemzar; Eli Lilly & Co, Indianapolis, IN) apoptosis in chemotherapy-resistant PDAC cells.10 Chemoradiotherapy with gemcitabine is frequently used to treat LAPC. Chemoradiotherapy can sometimes result in secondary resectability, which is the only potentially curative therapy, but this is typically achieved in only approximately 10% to 15% of treated patients.11
The first report of an antitumoral activity of an HIV protease inhibitor (HPI) was published in 1999 as a case report,12 and the potential use of these drugs in oncology has been recently discussed.13 Utilization of nelfinavir (Viracept; F. Hoffmann-La Roche Ltd, Basel, Switzerland) as an anticancer agent would have the advantage that the drug has already been used for more than a decade in the clinic and its safety profile is well defined. The radiosensitizing effect of HPIs has been well characterized in vitro and in vivo and has been linked to the inhibition of phospho-Akt.14,15
On the basis of this preclinical evidence, we undertook a phase I dose-escalation study of nelfinavir combined with chemoradiotherapy in patients with borderline resectable and unresectable PDAC. Because the standard dose of nelfinavir for HIV patients is known to be safe and does inhibit the phosphorylation of Akt,14 we chose to study two dose levels of gemcitabine. This drug is known to have a narrow therapeutic window in conjunction with radiotherapy.16 The primary end point was to evaluate the toxicity of nelfinavir in concurrent use with chemoradiotherapy. Combined positron emission tomography (PET) and computed tomography (CT) imaging and CA19-9 tumor marker course were used to evaluate the response of the patients. Six patients successfully underwent a margin-negative resection (R0) after multimodal therapy. Another patient underwent an attempted resection but was found to be unresectable.
PATIENTS AND METHODS
Eligibility
Adult patients with histologically proven PDAC were enrolled onto this prospective study. Resectability was based on vessel involvement.17 The National Comprehensive Cancer Network Clinical Practice Guidelines V.I.2008, which were not available at the time of initiation of the study, have been applied retrospectively to define resectability status18 for the purpose of this analysis. Additional inclusion criteria were a Karnofsky performance status ≥ 70% and sufficient hematologic and renal function. The Institutional Review Board of the University of Erlangen approved the trial, which was in accordance with the precepts of the Declaration of Helsinki. Approval by the German Federal Drug Administration was granted, and informed consent was obtained. Patients who developed metastasis during therapy switched to systemic chemotherapy.
Treatment and Evaluation of the Effect of Nelfinavir on Leukocyte Phospho-Akt
Oral administration of nelfinavir at the daily standard dose of antiretroviral treatment (2 × 1,250 mg) was added to chemoradiotherapy (Fig 1). This dose of nelfinavir has been shown to reliably inhibit Akt phosphorylation.14 A twice-daily dosage was chosen over a thrice-daily dosage because a higher 24-hour area under the curve was achieved. Nelfinavir was started 3 days before radiation and was continued throughout the last day of radiotherapy. Two dose levels (DLs) of gemcitabine with six patients each were studied. Gemcitabine is known for a narrow therapeutic window with chemoradiotherapy, and a dose of 300 mg/m2 concurrently with radiation with 30 mg/m2 of cisplatin (Cis-GRY; GRY-Pharma, Kirchzarten, Germany) had previously been determined to be safe.16,19 For DL1, a dose of gemcitabine 200 mg/m2 was chosen, with escalation to gemcitabine 300 mg/m2 in DL2. Chemotherapy was administered on days 1, 8, 22, and 29. Three-dimensional treatment planning was used to ensure uniform dosimetry (Appendix, online only).20
Treatment schedule for patients with locally advanced pancreatic cancer. Each arrow in the upper half represents one fraction of radiotherapy delivered. XRT, radiotherapy; Cis, cisplatin.
Patients' blood was drawn before the first dose of nelfinavir and during radiotherapy to examine the effect of nelfinavir on leukocyte Ser473-phospho-Akt levels. Phospho-Akt levels were determined by immunoblotting (antibody from Cell Signaling, Danvers, MA) after Ficoll density gradient centrifugation. Quantification of immunoreactions was provided using an ImageJ-based density analysis (http://rsb.info.nih.gov/ij/). The measured immunoblot density (D) of phopho-Akt/β-actin during treatment with nelfinavir (Dnfv) was normalized to pretreatment (Dpre) phopho-Akt/β-actin levels as follows: (Dnfv phospho-Akt/Dnfv β-actin)/(Dpre phospho-Akt/Dpre β-actin).
Toxicity Evaluation, Imaging, and Response Assessment
Dose-limiting toxicities (DLTs) were defined by the National Cancer Institute's Common Terminology Criteria for Adverse Events version 3.0 and the Radiation Therapy Oncology Group/European Organisation for Research and Treatment of Cancer recommendations for grading of acute toxic effects of radiotherapy. Maximum-tolerated dose (MTD) was defined as DLT occurring in one of three or two of six patients. DLT was defined as grade 4 hematologic toxicity or nonhematologic toxicities of ≥ grade 3 during treatment or within 4 weeks after completion unless clearly attributable to other clinical causes (eg, stent occlusion). The MTD was defined as the dose of gemcitabine associated with DLT occurring in 33% of the patients treated at the same DL, and the recommended phase II dose was defined as the DL below the MTD.
Before treatment and 6 weeks after completion of chemoradiotherapy, all patients underwent PET/CT at controlled blood glucose levels (< 200 mg/dL) on a dedicated combined PET-CT simulator (Biograph; Siemens, Berlin, Germany). CT response assessment was performed using the Response Evaluation Criteria in Solid Tumors.21 For each PET study, the maximum standardized uptake value of the primary cancer was measured. The maximum pixel standardized uptake value within a region of interest encompassing the tumor was used. CA19-9 (U/mL) was measured, and the percent change at restaging was calculated.
Surgical Resection and Pathologic Evaluation
Patients with tumors that were judged to be resectable on restaging were operated on 6 to 7 weeks after chemoradiotherapy by partial duodenopancreatectomy or left pancreatic resection. Resection specimens were analyzed histologically to detect positive margins (R1 resection).
RESULTS
Patient Characteristics and Lymphocyte Phospho-Akt Inhibition
Twelve patients (six per DL) were enrolled onto the study between August 2006 and February 2007, and their characteristics are listed in Table 1. Nelfinavir reduced Ser473-phospho-Akt levels in the leukocytes of assessable patients, as shown in Figure 2.
Determination of the Ser473-phosphorylation status of Akt before treatment with nelfinavir and during treatment in peripheral-blood mononuclear cells in eight of 10 patients who completed therapy. β-actin was used to control for loading on Western blots. Quantification was related to β-actin and performed as described in Patients and Methods.
Clinicopathologic Characteristics for 12 Patients Receiving Nelfinavir and CRT for Locally Advanced Pancreatic Cancer
Toxicities
One patient (DL1) discontinued treatment after week 2 because of reactivated, previously undisclosed clinical depression, which was an exclusion criterion. Metastasis-related acute cholecystitis in another patient (DL2) caused treatment to be aborted after week 3 in favor of palliative chemotherapy. Both patients died 6 months after diagnosis from progressive disease. The other 10 patients completed chemoradiotherapy as planned. No dose reductions of gemcitabine or cisplatin were necessary. Mean nelfinavir intake was 94% at DL1 and 96% at DL2. One patient at each DL had reduced nelfinavir intake because of noncompliance (82%) and loss of appetite (87%). The frequencies of grade 3 and 4 toxicities observed during chemoradiotherapy are listed in Table 2. Observed hematologic toxicities reached grade 3 and consisted mostly of leukopenia. Patients in DL2 did not experience more severe hematologic toxicity, and no dose-limiting hematotoxicities (grade 4) were observed.
Acute Toxicity
Nonhematologic toxicities also did not occur at a higher frequency at DL2. Grade 3 nausea and vomiting occurred in one patient at each DL, and grade 3 upper abdominal toxicity occurred in one patient at DL1. This patient experienced grade 3 abdominal pain on days 10 through 12. He simultaneously had an increase of the bilirubin level with infection, both grade 3. These symptoms were caused by a biliary obstruction and necessitated hospitalization with endoscopic change of the biliary stent on day 12, resulting in a 2-day break from any oral medication including nelfinavir. Biliary obstruction was also observed in another patient at DL1, leading to combined grade 3 nausea/vomiting, bilirubinemia, and elevation of alkaline phosphatase and transaminases. These symptoms quickly disappeared after replacement of the biliary stent. Because the patient refused outpatient supportive parenteral feeding, which was an integral part of the protocol when necessary, a weight loss of more than 15% could not be prevented (grade 3 upper abdominal toxicity, no DLT). Post-therapeutic biliary obstruction caused a grade 4 elevation of transaminases and grade 3 infection in a patient at DL2. Values quickly returned to normal after stent replacement (no DLT). Grade 4 cholecystitis occurred in another patient at DL2 during week 3, and treatment had to be interrupted after 27 Gy. This patient was re-evaluated, and the diagnostic procedures revealed previously undetected peritoneal carcinomatosis, which was proven histologically by laparoscopy. Chemoradiotherapy was stopped, and the patient subsequently received palliative chemotherapy. Another patient at DL2 had grade 3 nausea and an episode of vomiting after chemotherapy. The symptoms responded to therapy with a 5-hydroxytryptamine-3 receptor antagonist.
None of the observed toxicities, including the patient with reactivated depression, seemed to be related to nelfinavir treatment. MTD was not reached at DL2 because, at this DL, no grade 4 hematotoxicity occurred, and DL2 was identified as the recommended phase II dose. Furthermore, grade 3 nonhematologic toxicities (nausea/vomiting and infection in one patient) were not found to be related to nelfinavir. No further dose escalation was performed because nelfinavir reliably inhibits phosphorylation of Akt at 2 × 1,250 mg/d, and the recommended phase II dose of chemoradiotherapy with gemcitabine and cisplatin has been established at 300 mg/m2.16
Response, Resections, and Follow-Up
Table 3 lists the responses in the 10 patients who completed therapy. Using Response Evaluation Criteria in Solid Tumors; partial CT responses were observed in five of 10 patients, and minor responses were observed in two of 10 patients (Fig 3). Compared with CT, responses were superior by PET in six patients. PET restaging was not performed in one patient because of a defect of the PET scanner. PET detected distant metastasis in one patient at restaging. Initial CA19-9 values were above normal (0 to 37 U/mL) in nine patients and after correction for normal values (CA19-9 value of 37 U/mL), eight of nine patients had a reduction to less than 12% of the normalized initial values. The one patient with distant metastasis had an increasing CA19-9. Response did not correlate with changes in Akt phosphorylation as quantified in Figure 2.
Response to therapy. (A) Pretherapeutic and (B) post-therapeutic positron emission tomography (PET)/computed tomography (CT) scans of a patient with a tumor of the pancreatic head (cT3N0) who underwent complete resection 6 weeks after chemoradiotherapy with pathohistologically proven tumor sterilization (ypT3N0R0, Breslin grade IIb22). This was graded as a complete response in fluorodeoxyglucose-PET (difference in standard uptake values = 2.7). The green scale on the left margin indicates centimeters. (C) Sections stained with hematoxylin and eosin obtained by CT-guided core biopsy from a patient with a cT3N1G2 tumor (arrows) of the pancreatic head at ×10 magnification. (D) Resection specimen with more than 90% tumor regression [ypT3N1(1/21)R0, Breslin grade III] and only a few tumor cells remaining (arrow) at ×10 magnification. (E) Endoscopic biopsy at ×10 magnification from a patient with a cT3N1 tumor of the pancreatic head with (F) complete regression (ypT0N0, Breslin grade IV) at resection at ×4 magnification (insert at ×10 magnification). i, islet cells; s, stroma.
Response to Therapy, Resections, and Further Course of Disease
Six of 10 patients underwent laparotomy and margin-negative (R0) resection 6 weeks after therapy. Vascular involvement had been the obstacle for primary tumor resection in 11 patients, as specified in Table 3, and a tumor of the pancreatic tail (maximum diameter = 9 cm) was not resected at diagnosis because of the combination of large tumor size, infiltration of the kidney and spleen, and enlarged lymph nodes. Postsurgical tumor response grading according to Breslin22 revealed a high degree of destruction of the tumor cells, including one complete pathologic remission (Table 3).
At the time of analysis (October 2007), all fully treated patients were alive with 7 to 13 months of overall survival time. All patients with resected tumors were free from disease, except the patient with a large tumor of the pancreatic tail who was diagnosed with peritoneal carcinomatosis 4 months after resection. This patient subsequently received palliative chemotherapy. Adjuvant or additive chemotherapy was administered to all patients except one who refused adjuvant therapy.
DISCUSSION
The present report demonstrates the feasibility of combining nelfinavir with chemoradiotherapy for patients with borderline resectable or locally advanced PDAC. To our knowledge, this is the first clinical report evaluating the use of an HPI in oncology. We used a study design that was based on the dose of nelfinavir used in the treatment of HIV-positive patients. There was evidence of efficacy of this regimen from the responses observed.
Dose escalation of gemcitabine was chosen because gemcitabine can induce hematotoxicity in chemoradiotherapy.16 Higher grade nausea and vomiting is occasionally observed after cisplatin chemoradiotherapy. Other chemoradiotherapy regimens using higher doses of cisplatin have reported much higher rates of grade 3 nausea/vomiting than this study.23 Adding nelfinavir to chemoradiotherapy was well tolerated in this trial, with no toxicity greater than that expected with chemoradiotherapy alone, where higher grade leukopenia and thrombocytopenia are dose limiting.16,19,24 In this trial, no cases of grade 4 leukopenia were observed, and grade 3 leukopenia occurred in three patients at DL1 and one patient at DL2. Two patients had grade 3 thrombocytopenia without hemorrhage, one at each DL. Treatment-related grade 3 nausea/vomiting was seen in two patients.
Occluded biliary stents were the cause of non–treatment-related grade 3 and 4 nonhematologic toxicities in three of 12 patients and quickly resolved after stent exchange. By comparison, it has previously been reported that 15 of 101 patients with stents needed stent exchange as a result of occlusion during chemoradiotherapy. This successfully prevented uncontrolled biliary sepsis, hepatic abscesses, and stent-related death.25 Frequency and severity of cholestatic events were not higher than expected in this study. Disease-related grade 4 acute cholecystitis was diagnosed in a patient who had pathohistologic confirmation of peritoneal metastasis during cholecystectomy. In summary, the addition of nelfinavir to chemoradiotherapy did not cause increased toxicity in this trial.
Five of 10 patients had a significant partial CT response in this trial, and six of 10 patients underwent resection. In contrast to other tumors, the size of pancreatic tumors is rarely decreased after chemoradiotherapy26 because of the pronounced desmoplastic reaction of this tumor.27,28 The reported CT responses after chemoradiotherapy are typically around 30%.3,16 Response was also measured with [18F]fluorodeoxyglucose-PET, and consistent with a previous report, a higher number of responses was found by PET scanning compared with CT scanning,26 with five complete responses documented by this method. Therefore, PET scanning might be better able to discriminate tumor tissue from fibrosis. CA19-9 tumor marker levels decreased after therapy in eight of nine assessable patients (to a median of 12% of initial values). The only patient with an increasing CA19-9 level at restaging had distant disease. Progressive disease has previously been described to correlate with an increase in CA19-9 levels after chemoradiotherapy in PDAC, whereas decreasing values were less predictive for clinical response, perhaps because of the failure to correlate with the absolute initial value.29 The observed PET/CT and CA19-9 responses support the hypothesis that nelfinavir may enhance chemoradiotherapy in borderline or unresectable PDAC. Secondary complete resections in six of 10 tumors, with one sterilized tumor, in this trial are a possible indicator of a good local effect of nelfinavir with chemoradiotherapy. Historical series of chemoradiotherapy for LAPC report secondary resectability in the range of 15% to 20% of the patients.11 Within our own center, the resectability rate was 20% after chemoradiotherapy in patients with the same eligibility criteria as in this study.30
Given the results obtained in this trial, the potential radiosensitizing and antitumor mechanisms of HPIs are of high interest. The radiosensitizing effects of HIV inhibitors have been attributed to dephosphorylation of Akt.14 In xenograft models, doses of nelfinavir comparable with the therapeutic levels achieved in HIV patients were shown to inhibit activation of Akt, and radioresistance could be reverted by nelfinavir treatment in clonogenic assays.14 Subsequent investigations proposed a mechanism for decreased phosphorylation of Akt. Proteasome inhibition by the drug is proposed to trigger the unfolded protein response. This in turn induces the GADD34/PP1 complex, which dephosphorylates Akt.15 Consistent with this hypothesis, proteasome inhibition has been tested as a radiosensitizing strategy.31 Recently, a phase I study was activated in LAPC combining chemoradiotherapy with a proteasome inhibitor.32
The antitumor effects of HPIs have been attributed to several effects.33 Among the proposed molecular tumor targets of HPIs are the proteasome, matrix metalloproteases, and integrins. Non–Akt-related effects of proteasome inhibition include elevated endoplasmic reticulum stress, increased protein turnover, reduced clearance of misfolded proteins, apoptosis induction, degradation of tumor-suppressor gene products, altered function of cyclin-dependent kinase inhibitors and of nuclear factor-κ B, and blockage of the inflammatory response. HIV protease–induced matrix metalloprotease inhibition also has immunomodulatory effects as well as antiangiogenic effects. A recent study reported that the sensitization of endothelial cells to radiation by nelfinavir mediated upregulation of apoptosis in this cell type, and a significant reduction of tumor vessel density after combined treatment was observed.34 Other antitumor mechanisms of HPIs include the inhibition of inflammatory cytokines, immune modulation, induction of apoptosis, and induction of autophagy.13 Taken together, the specific radiosensitizing effects of HPI have been partially identified but will require further investigation.
Future research will have to decipher the relative contribution of the supposed mechanisms of action and identify the optimal scheduling of combined HPI and (chemo)radiation. The identified actions of HPIs on tumor cell radiation survival, angiogenesis, and tumor immunology will require the study of the effect of HPIs on tumors in the context of their microenvironment. In light of the fact that the HPIs are well tolerated, it will be possible to complement these investigations by clinical trials. To date, phase I trials combining HPI and radiation have been activated in lung cancer and rectal cancer.
In conclusion, this report demonstrates that the use of nelfinavir with chemoradiotherapy is safe in locally advanced PDAC because no added toxicity was observed. The observed responses and the favorable tumor response grading of resected tumors after therapy indicate the potential for efficacy. Therefore, this combination merits further assessment, and a phase II study in PDAC is being prepared.
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
Although all authors completed the disclosure declaration, the following author(s) indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a “U” are those for which no compensation was received; those relationships marked with a “C” were compensated. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.
Employment or Leadership Position: None Consultant or Advisory Role: None Stock Ownership: None Honoraria: None Research Funding: Thomas B. Brunner, Roche Expert Testimony: None Other Remuneration: None
AUTHOR CONTRIBUTIONS
Conception and design: Thomas B. Brunner, Gerhard G. Grabenbauer, W. Gillies McKenna
Financial support: Rolf Sauer
Administrative support: Gerhard G. Grabenbauer, Marga Lang-Welzenbach, Rolf Sauer, W. Gillies McKenna
Provision of study materials or patients: Thomas B. Brunner, Matthias Geiger, Gerhard G. Grabenbauer, Tine S. Mantoni, Rolf Sauer, Werner Hohenberger
Collection and assembly of data: Thomas B. Brunner, Matthias Geiger, Marga Lang-Welzenbach, Tine S. Mantoni, Alexander Cavallaro
Data analysis and interpretation: Thomas B. Brunner, Matthias Geiger, Alexander Cavallaro
Manuscript writing: Thomas B. Brunner, W. Gillies McKenna
Final approval of manuscript: Thomas B. Brunner, Matthias Geiger, Gerhard G. Grabenbauer, Marga Lang-Welzenbach, Tine S. Mantoni, Alexander Cavallaro, Rolf Sauer, Werner Hohenberger, W. Gillies McKenna
Appendix: Methods and Materials
Details of radiotherapy.
Radiotherapy was administered at 1.8 Gy per fraction to a total of 50.4 Gy including the regional lymphatics, followed by a boost to 59.4 Gy to the tumor. Regional lymphatics included superior, inferior, ventral and dorsal pancreatico-duodenal, pyloric, celiac, proximal mesenteral nodes as well as nodes alongside the choledochal duct for pancreatic head tumors.20 In addition, splenic lymph nodes were included in the patient with a tumor of the pancreatic tail.
Details of supportive measures.
If necessary, parenteral nutrition was provided to guarantee a sufficient supply of a minimum of 1,500 kcal/24 h via a Port-a-cath. This supplement included carbohydrates, proteins, and fats. Supplemental nutrition was provided if a pretherapeutic loss of weight > 10% was observed or if loss of weight > 5% occurred during chemoradiation. Chemotherapy was provided to patients both with or without resection after the course of chemoradiation. It consisted of gemcitabine alone (1,000 mg/m2 on days 1, 8, and 15, every 29 days) and was started 4 weeks after surgery or 6 weeks after completion of chemoradiation if no resection was possible.
Acknowledgments
We thank Kerstin Amann, MD, from the Institute of Pathology at the University of Erlangen, who provided pathology advice and microscopy; and Eric J. Bernhard, PhD, for helpful support with editing the manuscript.
Footnotes
-
Supported by the University Hospitals of Erlangen and by Grants No. H3RMWX0 from the Medical Research Council (T.B.B., W.G.M.), H3RMGW0 from Cancer Research UK (W.G.M.), and 5PO1-CA-075138 from the National Cancer Institute (T.B.B., W.G.M.). T.B.B. received research funds from F. Hoffmann-La Roche Ltd, Basel, Switzerland.
-
Both T.B.B. and M.G. contributed equally to this work.
-
Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
-
Clinical trial information can be found for the following: EudraCT 2006-001766-17.
- Received November 12, 2007.
- Accepted February 19, 2008.
