Studies of intraoperative radiotherapy in carcinoma of the pancreas.

Intraoperative radiotherapy (IORT) involves the administration of therapeutic radiation to malignancies during surgical procedures. IORT permits high dose delivery to tumors with the simultaneous reduction of radiation exposure to normal tissues, which may be directly shielded or operatively mobilized from the treatment volume. IORT has been investigated in various intra-abdominal malignancies, including carcinoma of the pancreas. Techniques of IORT were initially developed in Japan during the 1970's. Reports of therapeutic benefit in some patients with unresectable pancreatic cancer encouraged further examinations by various institutions in the United States. Experiences at the Massachusetts General Hospital in the early 1980's suggested that IORT enhanced survival in selected patients with locally advanced but non-metastatic disease. However, subsequent investigations studies by a variety of institutions, including the Mayo Clinic, failed to establish any conclusive evidence that IORT significantly prolonged the survival enhancement of unresectable pancreatic cancer patients. A prospective multi-institutional study carried out by the Radiation Therapy Oncology Group (RTOG) showed an 8-month median survival, similar to conventional therapy and indicating that IORT failed to prolong survival. However, the RTOG did show that IORT rapidly and consistently palliated the severe viseral pain which often accompanied pancreatic cancer. By 1990, some institutions had explored IORT as an adjunct to pancreatectomy, in patients with resectable tumors. Studies typically involved highly selected uncontrolled patients but did suggest that IORT could enhance local disease control and, in some cases, overall survival, when performed in conjunction with pancreatic resection. IORT appeared to be most conspicuously beneficial when used with extended radical resections. A small prospectively randomized trial conducted at the National Cancer Institute showed significant improvement in local disease control in patients receiving IORT compared with patients receiving conventional external beam postoperative radiotherapy after resection. Current evidence suggests that IORT may have an important palliative role in patients with unresectable pancreatic cancers, ameliorating visceral pain and promoting local control of the primary tumor; however, IORT appears to have no significant effect on overall survival. For patients with resectable disease, especially patients with locally extensive tumor, IORT appears to have benefit in enhancing disease control and in some cases survival. It is reasonable to further explore the potential role of IORT in pancreatic cancer, especially as a component of multimodal therapy, since IORT's demonstrated enhancement of local control could be an important factor in eventual disease control.

Randomized prospective study of the benefit of adjuvant radiation therapy in the treatment of soft tissue sarcomas of the extremity.

PURPOSE: This randomized, prospective study assesses the impact of postoperative external-beam radiation therapy on local recurrence (LR), overall survival (OS), and quality of life after limb-sparing resection of extremity sarcomas. PATIENTS AND METHODS: Patients with extremity tumors and a limb-sparing surgical option were randomized to receive or not receive postoperative adjuvant external-beam radiotherapy. Patients with high-grade sarcomas received postoperative adjuvant chemotherapy whereas patients with low-grade sarcomas or locally aggressive nonmalignant tumors were randomized after surgery alone. RESULTS: Ninety-one patients with high-grade lesions were randomized; 47 to receive radiotherapy (XRT) and 44 to not receive XRT. With a median follow-up of 9.6 years, a highly significant decrease (P2 = .0028) in the probability of LR was seen with radiation, but no difference in OS was shown. Of 50 patients with low-grade lesions (24 randomized to resection alone and 26 to resection and postoperative XRT), there was also a lower probability of LR (P2 = .016) in patients receiving XRT, again, without a difference in OS. A concurrent quality-of-life study showed that extremity radiotherapy resulted in significantly worse limb strength, edema, and range of motion, but these deficits were often transient and had few measurable effects on activities of daily life or global quality of life. CONCLUSION: This study indicates that although postoperative external-beam radiotherapy is highly effective in preventing LRs, selected patients with extremity soft tissue sarcoma who have a low risk of LR may not require adjuvant XRT after limb-sparing surgery (LSS).

Intraoperative radiotherapy for pancreatic carcinoma. Experimental and clinical studies.

Intraoperative radiotherapy (IORT) is an innovative treatment approach that has been tested in several Phase I-II studies and small Phase III trials in patients with resectable and unresectable pancreatic cancer. The technical approach and dose guidelines for adjacent normal tissues have been established in a series of experiments using a large animal (canine) model. This article reviews the experimental and clinical studies of IORT in cancer of the pancreas.

A phase I/II study of dose-intense paclitaxel with cisplatin and cyclophosphamide as initial therapy of poor-prognosis advanced-stage epithelial ovarian cancer.

Epithelial ovarian cancer patients with bulky residual tumor have a poor response to therapy and limited survival. We investigated the addition of dose-intense paclitaxel to cisplatin and cyclophosphamide for patients with FIGO III/IV epithelial ovarian cancer. Paclitaxel dose was intensified from 135 to 250 mg/m2 and administered in combination with cisplatin at > or = 75 mg/m2 and cyclophosphamide at 750 mg/m2. Thirty-one of 36 patients (86%) and 25 (70%) had > or = 2 and > or = 3 cm residual disease after surgery, respectively. One-third had stage IV disease, and 80% had grade 3 tumors. The maximally tolerated doses (MTD) were paclitaxel at 250 mg/m2, cisplatin at 75 mg/m2, and cyclophosphamide at 750 mg/m2 on a 21-day cycle with G-CSF, 10 micrograms/kg/day. Administered dose intensity at the MTD was > or = 86%. Reversible grade 3 peripheral neuropathy occurred in 28% of patients and fever during neutropenia in 2/352 cycles (0.5%). The pathologic response rate is 36% with an additional 25% having minimal microscopic disease. Median progression-free and overall survivals for patients receiving paclitaxel at 250 mg/m2 at a median potential follow-up of 22 months have not been reached for the cohort nor for the > or = 3-cm subgroup. This regimen should be evaluated in a prospective, randomized clinical trial.

Cytologic effects of photodynamic therapy in body fluids.

Photodynamic therapy (PDT) has been used in phase I clinical trials at the National Institutes of Health for the treatment of malignancies disseminated within the peritoneal and pleural cavities. Review of records revealed 18 patients who were treated with PDT between April 1988-June 1993. Sixty-five pleural and peritoneal fluids, 22 pre- and 43 post-PDT, were available for evaluation. Mesothelial cell changes seen post-PDT included: increased nuclear-to-cytoplasmic ratios in 7/18 (39%), cytomegaly in 9/18 (50%), and multinucleation in 12/18 (67%), with Touton-like giant cells in 3/18 (17%). Additional changes noted post-PDT comprised histiocytic aggregates in 9/18 patients (50%), with granuloma-like clusters in 3/18 (17%), acute and chronic inflammation in 13/18 (72%), and eosinophilia in 8/18 (44%). Residual tumor was present in 7/18 (39%) patients post-PDT. In 2 patients with malignant mesothelioma, benign mesothelial cells with cytologic changes post-PDT were difficult to distinguish from malignant cells. Mesothelial cell changes following PDT, specifically increased nuclear-to-cytoplasmic ratios and cytomegaly, should be recognized to avert false-positive diagnoses of tumor. In patients with malignant mesothelioma, and less commonly with adenocarcinoma, benign mesothelial cells with changes secondary to PDT may be difficult to distinguish from tumor cells.

The use of transureteroureterostomy in the management of complex ureteral problems.

PURPOSE: Transureteroureterostomy has primarily been performed for benign disease in children with reflux or undergoing undiversion, and in adults with lower ureteral injury. We report the use of transureteroureterostomy for other than these traditional indications, including malignant disease. MATERIALS AND METHODS: Transureteroureterostomy was performed in 6 patients at the time of tumor resection to bypass large ureteral defects and in 4 as a secondary operation, usually associated with a ureteral leak after previous surgery. RESULTS: Complications related to transureteroureterostomy included 1 ureteral stricture and 1 ureteral leak. Good renal function was maintained with a mean followup of 77.9 months. CONCLUSIONS: Our experience suggests that transureteroureterostomy can be useful for urinary diversion when a segment of lower ureter is involved with malignant disease.

Tumors in dogs exposed to experimental intraoperative radiotherapy.

PURPOSE: The frequency of radiation-induced neoplasms was determined in dogs enrolled in the National Cancer Institute canine trials of intraoperative radiotherapy (IORT). METHODS AND MATERIALS: Twelve protocols assessing normal tissue response to IORT involved 238 dogs in a 15-year trial. Eighty-one dogs were followed for > 24 months postoperatively and were assessed for tumor development; 59 of these animals received IORT. RESULTS: Twelve tumors occurred in the 59 dogs receiving IORT. Nine were in the IORT portals and were considered to be radiation induced. No tumors occurred in 13 sham animals or in 9 animals treated with external beam radiotherapy alone. The frequency of radiation-induced malignancies in dogs receiving IORT was 15%, and was 25% in animals receiving > or = 25 Gy IORT. Frequency of all tumors, including spontaneous lesions, was 20%. CONCLUSIONS: Intraoperative radiotherapy contributed to a high frequency of sarcoma induction in these dogs. Unknown to date in humans involved in clinical trials of IORT, this potential complication should be looked for as long-term survivors are followed.

Radical reoperation for advanced pancreatic carcinoma.

The indications and outcomes of aggressive reoperation in patients referred to the National Cancer Institute (NCI) for protocol therapy of locally advanced pancreatic carcinoma were investigated. Twenty-nine patients referred to the NCI after exploration and determination of unresectability elsewhere were considered to have localized disease after a metastatic work-up. These patients were then entered onto NCI adjuvant therapy protocols and taken to exploratory laparotomy. Intraoperatively, patients underwent complete resection if possible; otherwise varying palliative surgical procedures were performed. Of the 29 patients, 16 underwent complete resection of their disease, and 13 were unresectable. Two patients suffered postoperative mortality. Disease-specific survival of the resected patients was significantly better than that of the unresectable patients (P < 0.01). The two long-term survivors (53 and > 109 months) underwent definitive surgery after a palliative procedure elsewhere. Complete resection of pancreatic carcinoma contributes to increased survival. The intraoperative definition of unresectability in pancreatic cancer varies with the degree of pancreatitis present, the surgical expertise of the surgeon, and the available ancillary services. Given the extremely grave prognosis of patients with unresectable pancreatic carcinoma, locally unresectable patients without peritoneal seeding of distant metastases at exploration should be considered for referral for protocol therapy to centers where expertise in radical surgery for pancreatic cancer exists.

Hemipelvectomy and intraoperative radiotherapy for bone and soft tissue sarcomas of the pelvic girdle.

Current treatment of locally advanced bone and soft tissue sarcomas of the pelvic girdle are associated with a high local and distant failure rate, and local tumor control after hemipelvectomy can be a significant problem. IORT has been used in conjunction with hemipelvectomy, both conventional (seven patients) and limb-sparing internal hemipelvectomy (one patient), in seven males and one female, median age 27 (range 24-57) years with locally extensive high grade bone (seven patients) or soft tissue (one patient) sarcomas. IORT (15-30 Gy, 8-16 MeV) was delivered to sacral resection margins and surrounding soft tissues considered likely to harbor microscopically residual disease. Four patients received 46-54 Gy postoperative radiotherapy in addition to IORT. During a median follow-up of 33 months (range 6-131 months) two patients developed a local recurrence (25%), and five patients distant metastases (62%). Three patients with pelvic girdle sarcomas remained free of tumor (37%) with a mean follow-up of 100 (range 49-131) months. There was no treatment-related mortality. Two patients developed radiation-induced necrosis of the coccyx (25%). On the basis of this preliminary experience, it appears that IORT may substantially help to control local recurrence and survival in patients with marginally resectable sarcomas of the pelvic girdle after hemipelvectomy. Since the majority of the patients die from metastatic disease, there is a need for adjuvant systemic treatment.

Clinical toxicity of peripheral nerve to intraoperative radiotherapy in a canine model.

PURPOSE: The clinical late effects of intraoperative radiotherapy (IORT) on peripheral nerve were investigated in a foxhound model. METHODS AND MATERIALS: Between 1982 and 1987, 40 animals underwent laparotomy with intraoperative radiotherapy of doses from 0-75 Gy administered to the right lumbosacral plexus. Subsequently, all animals were monitored closely and sacrificed to assess clinical effects to peripheral nerve. This analysis reports final clinical results of all animals, with follow-up to 5 years. RESULTS: All animals treated with > or = 25 Gy developed ipsilateral neuropathy. An inverse relationship was noted between intraoperative radiotherapy dose and time to neuropathy, with an effective dose for 50% paralysis (ED50) of 17.2 Gy. One of the animals treated with 15 Gy IORT developed paralysis, after a much longer latency than the other animals. CONCLUSIONS: Doses of 15 Gy delivered intraoperatively may be accompanied by peripheral neuropathy with long-term follow-up. This threshold is less than that reported with shorter follow-up. The value of ED50 determined here is in keeping with data from other animal trials, and from clinical trials in humans.

Evolving strategies for the treatment of adenocarcinoma of the pancreas. A review.

Adenocarcinoma of the pancreas has an incidence of only 0.01%, yet is the fourth leading cause of cancer death for American men and women. Despite this dismal outlook, new strategies for staging and therapy for pancreatic cancer have emerged over the last few years. Laparoscopy with cytologic evaluation of peritoneal washings, and more recently, although still investigational, endoscopic and intracorporeal ultrasonography have provided more detailed staging information. The result of improved staging is earlier, more accurate selection of treatment most appropriate for stage of disease. For those patients with clinically localized disease, laparotomy with an attempt at resection is indicated, particularly with the recent trend in declining morbidity and operative mortality the recent trend in declining morbidity and operative mortality associated with pancreatectomy. With clinically unresectable disease, patients may potentially be spared the morbidity of laparotomy. Advances in therapeutic endoscopic and percutaneous manipulation of the obstructed biliary tree have provided an alternative to surgery and improved quality of life for patients with abbreviated life spans. Gastroduodenal obstruction has traditionally been managed by laparotomy, although with improved technology and surgical skill, a laparoscopic approach may become standard. Because even at presentation pancreatic cancer is rarely a localized process but is a disseminated disease, surgery alone is unlikely to increase survival rates in the absence of adjuvant therapies. Present and future strategies for treatment include the addition of neoadjuvant regimens and adjuvant modalities including intraoperative radiation, photodynamic therapy, intraperitoneal therapies, and pancreatic and splanchnic perfusion. Clearly, the greatest strides in treatment of pancreatic cancer will come with development of new agents with significantly greater antitumor efficacy.

Iatrogenic transplantation of osteosarcoma.

We report a case in which a patient having curettage and resection of a presumed benign lesion of the tibia (later recognized as osteosarcoma) had probable iatrogenic transplantation of tumor to the contralateral iliac crest, which had served as a donor site for bone chips used to pack the tibial lesion. The patient later had above-knee amputation for the primary tumor and eventually required contralateral hemipelvectomy when tumor developed in the iliac crest donor site. We discuss the literature of tumor transplantation and seeding in operative settings and stress the clinical importance of avoiding possible tumor contamination of operative fields by meticulous instrument changes and by isolation of multiple surgical fields.

Hemipelvectomy for soft tissue sarcoma: clinical results in fifty-three patients.

The management of high-grade extremity sarcomas typically combines local surgical excision with some form of adjuvant therapy. However, some patients with large proximal lesions present with no local resection options, and high radical amputation is often the only way to accomplish complete tumour excision. A retrospective review was made of the clinical courses of 53 patients with high-grade sarcomas of the lower extremity who underwent hemipelvectomy with curative intent at the National Institutes of Health between 1975 and 1989. Hemipelvectomy provided excellent local control, with no local recurrences. Although 60% of the patients experienced post-operative complications, the operative mortality was zero. Ten years after hemipelvectomy, the actuarial disease-free rate was 25% and the survival rate was 35%. There were no statistically significant survival differences when patients were stratified according to tumour size, perioperative blood transfusion requirements or the use of adjuvant chemotherapy. We conclude that hemipelvectomy provides excellent local control and a reasonable cure rate for patients with locally unresectable soft tissue sarcomas of the lower extremity.

Tolerance of the bladder to intraoperative radiation in a canine model: a five-year follow-up.

PURPOSE: Late effects of intraoperative radiation therapy (IORT) on bladder were investigated in a canine model. METHODS AND MATERIALS: After laporatomy and cystotomy in adult female foxhounds weighing 25-35 kg, 12 MeV electrons were delivered intraoperatively to a 5 cm circular bladder field which included the trigone and both uretero-vesicle junctions. Each animal received doses of 0, 20, 25, 30, 35, or 40 Gy. All the dogs were followed 5 years postoperatively. An unoperated dog receiving no surgery or radiation treatment was followed as a control. Close clinical monitoring was performed with regular cystometrics and intravenous pyelography. Animals were killed as scheduled with complete necropsies, including histopathology, with special attention to genitourinary structures. RESULTS: There were no acute or late bladder complications detected clinically in any animal. The dog receiving 30 Gy IORT developed rhabdomyosarcoma in the treatment field at 58 months. On follow-up testing over 5 years, there was no loss of bladder contractility on cystometry, and mild changes in the ureters on intravenous pyleography when animals receiving IORT were compared with baseline pretreatment values or with control animals. Histologically, a difference was evident between irradiated and unirradiated animals, but the changes were not clearly dose-related. CONCLUSION: Intraoperative radiation therapy may by safely delivered to the canine bladder with few acute or chronic complications. It is an approach which has potential for clinical use and should continue to be explored in human clinical trials.

Experimental and clinical studies of intraoperative radiation therapy.

Intraoperative radiation therapy (IORT) is an innovative treatment modality that has recently been given considerable attention as an approach toward controlling various locally advanced cancers. IORT involves surgical extirpation or debulking of the malignant lesion and the delivery of a large single dose of radiation to the tumor bed or to residual disease. This strategy allows for a theoretical enhancement of the therapeutic effect of radiation for three reasons: (1) the biologic effectiveness of a single large radiation dose is higher than for the same dose given in a fractionated regimen; (2) the dose of radiation is precisely given to the area at greatest risk of tumor recurrence (or persistence); and (3) irradiation of dose-limiting normal tissues may be avoided by operative mobilization of the tissues from the treatment volume by customized lead shielding of anatomically fixed structures or by judicious choice of electron beam energies or use of a bolus to limit dose to deep structures. Electrons are generally used for IORT because of sharp dose falloff. This avoids potential toxic effects to normal structures that may lie deep to the treatment volume. Conventional external beam photon radiation therapy (EBRT) allows less accurate tumor volume delineation (even with sophisticated treatment planning technique) and dose limitations necessitated by normal tissues incidentally in the treatment volume. A considerable amount of experimental and clinical data are available on the acute and late effects of IORT on normal tissues. Dose tolerances of many organs have been described in large animal models, and clinical toxicities are evident in several trials. Clinical IORT treatments are provided in more than 250 U.S. and foreign centers at the present time. Given the current interest in IORT, this monograph will review IORT methods and experimental and clinical results with emphasis on its present and future role for locally advanced cancers.

Effects of intraoperative radiotherapy on vascular grafts in a canine model.

PURPOSE: The effects of intraoperative radiotherapy +/- external beam radiotherapy on prosthetic vascular grafts were investigated in a canine model. METHODS AND MATERIALS: In 1986 and 1987, 30 adult beagles underwent laparotomy with transection and segmental resection of the infrarenal aorta followed by immediate reconstruction with a prosthetic graft. Intraoperative radiotherapy at varying doses from 0-30 Gy was then administered to all animals. Half of the dogs received 36 Gy external beam radiotherapy in 10 fractions postoperatively. Animals were sacrificed and necropsied at predetermined intervals and as clinically indicated to assess early (< or = 6 months) and late (> 6 months) effects to the vascular graft and surrounding normal tissue. RESULTS: Histopathologic analyses of irradiated vascular structures were performed and correlations were made with the clinical outcome. The most frequent early clinical toxicity was graft thrombosis, occurring in 7 of 10 animals followed for < or = 6 months. Early graft thrombus formation appeared unrelated to radiotherapy dose and probably represented a technical surgical complication. Anastomotic stenosis of varying severity occurred in most animals followed > 6 months. Late (> 6 months) graft stenosis was correlated with intraoperative radiotherapy dose. At < or = 20 Gy of intraoperative irradiation, 3 of 14 animals developed late graft occlusion; at > 25 Gy, five of six animals developed late occlusion. On histopathologic review, increasing intraoperative dose and increasing total radiotherapy dose (intraoperative+external beam) appeared to correspond with increasing severity of graft changes seen after 6 months of follow-up. CONCLUSIONS: Thrombus formation is a frequent early complication of vascular graft placement of the infrarenal aorta in our beagle dog model. Intraoperative doses up to 20 Gy appear to contribute minimally to late graft occlusion, while doses > or = 25 Gy contribute to late occlusion with high likelihood. Both intraoperative dose and total radiotherapy dose correlated with late graft occlusion, and with histopathologic changes in the graft and anastomoses.

Postoperative complications in patients receiving suramin therapy.

BACKGROUND: Suramin is an antiparasitic agent that is currently being evaluated for antineoplastic activity. Documented toxicities of suramin include adrenal and renal insufficiency, coagulation factor abnormalities, immunosuppression, and polyneuropathy. These adverse effects have potential for contributing to postoperative morbidity in surgical patients. Because no experience with suramin has been reported in the surgical literature, this 5-year retrospective review of postoperative complications in patients receiving suramin was performed. METHODS: From a review of 171 charts, 14 patients were identified who had undergone a major surgical procedure either while receiving intravenous suramin or within 1 year after its administration. Primary diagnoses included prostate cancer (six), lymphoma (four), ovarian cancer (two), colon cancer (one), and glioblastoma (one). All patients received replacement dose hydrocortisone at the initiation of suramin therapy and thereafter. RESULTS: Eighteen major surgical procedures were performed with 18 complications occurring in five patients. The predominant complications encountered were hemorrhage (five), impaired wound healing (three), and bowel dysmotility (two). A highly significant relationship existed between the incidence of complications and interval from completion of suramin therapy to the time of operation (p < 0.0005), with 17 of the 18 morbidities occurring within the first month. The length of operation (p < 0.05) and amount of blood transfused during the procedure were related to postoperative morbidity (p < 0.05). No other factors evaluated were correlated to complications. CONCLUSIONS: This experience suggests the avoidance of elective procedures during the first month after suramin therapy and a heightened awareness of the potential for bleeding and wound healing problems in patients receiving suramin who do require an emergent procedure.

Late effects of intraoperative radiation therapy on retroperitoneal tissues, intestine, and bile duct in a large animal model.

PURPOSE: The late histopathological effects of intraoperative radiotherapy (IORT) on retroperitoneal tissues, intestine, and bile duct were investigated in dogs. METHODS AND MATERIALS: Fourteen adult foxhounds were subjected to laparotomy and varying doses (0-45 Gy) of IORT (11 MeV electrons) delivered to retroperitoneal tissues including the great vessels and ureters, to a loop of defunctionalized small bowel, or to the extrahepatic bile duct. One control animal received an aortic transection and reanastomosis at the time of laparotomy; another control received laparotomy alone. This paper describes the late effects of single-fraction IORT occurring 3-5 years following treatment. RESULTS AND CONCLUSION: Dogs receiving IORT to the retroperitoneum through a 4 x 15 cm portal showed few gross or histologic abnormalities at 20 Gy. At doses ranging from 30-45 Gy, radiation changes in normal tissues were consistently observed. Retroperitoneal fibrosis with encasement of the ureters and great vessels developed at doses > or = 30 Gy. Radiation changes were present in the aorta and vena cava at doses > or = 40 Gy. A 30 Gy dog developed an in-field malignant osteosarcoma at 3 years which invaded the vertebral column and compressed the spinal cord. A 40 Gy animal developed obstruction of the right ureter with fatal septic hydronephrosis at 4 years. Animals receiving IORT through a 5 cm IORT portal to an upper abdominal field which included a defunctionalized loop of small bowel, showed a few gross or histologic abnormalities at a dose of 20 Gy. At 30 Gy, hyaline degeneration of the intestinal muscularis layer of the bowel occurred. At a dose of 45 Gy, internal intestinal fistulae developed. One 30 Gy animal developed right ureteral obstruction and hydronephrosis at 5 years. A dog receiving 30 Gy IORT through a 5 cm portal to the extrahepatic bile duct showed diffuse fibrosis through the gastroduodenal ligament. These canine studies contribute to the area of late tissue tolerance to IORT.

Patterns of disease recurrence following definitive therapy of adenocarcinoma of the pancreas using surgery and adjuvant radiotherapy:correlations of a clinical trial.

PURPOSE: Between 1980 and 1984, 26 patients with resectable adenocarcinoma of the pancreatic head were enrolled in a National Cancer Institute protocol evaluating intraoperative radiotherapy vs. standard therapy. METHODS AND MATERIALS: After complete excision of their lesions, patients were observed (Stage I), or randomized to intraoperative radiotherapy vs. external beam radiotherapy (Stages II-IV). The intraoperative dose was 20 Gy in a single fraction using 9-20 MeV electrons. The external beam radiotherapy schema involved daily 150-180 cGy fractions to 45-55 Gy in 5-6 weeks. Chemotherapy was not used for primary disease but was administered off protocol for recurrent disease. Median potential followup for the trial was > 9 years, with a median patient survival of 18 months. Perioperative mortality was 27% (7 patients). Of the remaining 19 patients, one remains alive and without evidence of disease 9 years post-therapy. Twelve patients underwent autopsy and 2 required antemortem laparotomy; histopathologic evidence of disease recurrence was analyzed. RESULTS: Of 15 patients evaluable for intra-abdominal control, 7 (47%) suffered local recurrences and 7 (47%) failed regionally, with 5 patients (35%) failing in both areas. Five patients (35%) developed peritoneal seeding. Of 13 patients evaluable for systemic disease, 8 (62%) suffered distant failure. There were no differences in outcome between intraoperative or external beam radiotherapy or observation in this subset of patients. CONCLUSION: This report is among the most rigorous descriptions of patterns of failure after resection of pancreatic cancer. It appears clear that advances in local control of this disease are unlikely to translate into increased survival in the absence of improved systemic therapy.

Lymph node involvement and pancreatic resection: correlation with prognosis and local disease control in a clinical trial.

In a prospectively randomized trial evaluating pancreatic resection with adjuvant radiotherapy (intraoperative radiotherapy [IORT] vs. external beam radiotherapy [EBRT]), lymph nodal involvement was examined and correlated with outcome. Twenty-six patients underwent pancreatic resection and received either IORT or EBRT (Stages II-IV). Patients who were stage I received surgery alone. Regional nodal metastases were present in 15 of 26 (57%) patients. Seven patients suffered treatment-related mortality. Survival, mortality, and morbidity were unaffected by the type of radiotherapy. The survival of patients with negative nodes (median survival 24 months, range 10 to > 109) appeared superior to the survival of patients with nodal involvement (median survival 11.5 months; range 4-39). Even in patients with locally advanced disease extending into extrapancreatic tissues, two node-negative patients appeared to survive longer (12 and 53 months) than 10 node-positive patients with similarly extensive local disease (median survival 11.5 months; range 4-39). Local disease control, however, appeared to be independent of nodal involvement, with eventual local recurrences in 6 of 8 node-negative patients and in 4 of 7 node-positive patients who were evaluable for local disease control by autopsy or by antemortem laparotomy.