Pathway Mutations are Associated with Clinical Outcomes in Localized Pancreatic Cancer Treated with Neoadjuvant Chemoradiation Followed by Surgery

PURPOSE/OBJECTIVE(S)

The purpose of this study was to determine if mutations in biological pathways are associated with clinical outcomes in patients with localized pancreatic cancer who undergo neoadjuvant chemoradiation followed by surgical resection.

MATERIALS/METHODS

Patients treated with neoadjuvant chemoradiation followed by oncologic resection from 2015-2019 who also underwent next generation sequencing (NGS) of the primary tumor were included in this retrospective analysis. NGS was done using either Foundation One (n = 20), in-house Solid Tumor Panel (n = 121), or Tempus XT (n = 1). Genes were included in pathway analysis if at least one patient harbored a mutation in the gene. Pathways were defined from the Molecular Signatures Database Hallmark, KEGG, and Reactome gene sets. A pathway was deemed mutated if at least one gene within the pathway was mutated. Univariable Cox regression was performed to determine the association between pathway mutation status and overall survival (OS) as well as progression-free survival (PFS).

RESULTS

In total, 142 patients met criteria for study inclusion. For pathway analysis, 329 genes met inclusion criteria. Patients were typically treated with neoadjuvant chemotherapy (either 5-fluorouracil-based or gemcitabine-based) followed by radiation. Patients received SBRT (n = 104, most commonly 33 Gy in 5 fractions) or conventionally fractionated radiation (n = 38, most commonly 50.4 Gy in 28 fractions). For clinical variables, worse OS was significantly associated with T stage (p = 0.036), N stage (p = 0.044), and lymphovascular invasion (LVI, p = 0.011); worse PFS was significantly associated with T stage (p = 0.0008), N stage (p = 0.022), LVI (p = 0.026), and conventional RT (p = 0.007). Mutations in major pathways were associated with worse OS, notably hedgehog signaling (p = 0.001), chromatin modifying enzymes (p = 0.002), WNT/beta-catenin signaling (p = 0.005), mismatch repair (0.006), E2F targets (p = 0.008), FLT signaling (p = 0.012), VEGF signaling (0.025), innate immune system (p = 0.026), and NOTCH signaling (p = 0.029). Pathway mutations associated with worse PFS included mismatch repair (p = 0.007) and hedgehog signaling (p = 0.013).

CONCLUSION

For pancreatic cancer patients that undergo neoadjuvant chemoradiation followed by oncologic resection of the primary tumor, mutations in key biological pathways are associated with OS and PFS. Characterizing the importance of common pathway mutations may become increasingly valuable to help categorize less commonly mutated genes assayed by NGS.

Characterization of DNA Damage Response-Associated Somatic Mutations in Borderline Resectable and Locally Advanced Pancreatic Cancer

PURPOSE/OBJECTIVE(S)

The role of radiation for pancreatic cancer remains controversial, with recent studies showing conflicting results, highlighting the need to develop biomarkers of radiation response. Despite its potential utility in predicting radiosensitivity, the landscape of somatic mutations in borderline resectable pancreatic cancer (BRPC) and locally advanced pancreatic cancer (LAPC), as related to DNA damage response (DDR), has not been well characterized. This study aimed to characterize the frequency of such mutations in a cohort of patients with BRPC/LAPC treated with neoadjuvant chemotherapy and stereotactic body radiotherapy (SBRT).

MATERIALS/METHODS

Mutational data was collected from patients with BRPC/LAPC treated at a single institution with neoadjuvant chemotherapy and SBRT, followed by surgical resection from 2016-2021. Chemotherapy consisted of modified FOLFIRINOX or gemcitabine/nab-paclitaxel, and patients were treated with SBRT in 33 Gy in 5 fractions. Genomic data was obtained from either endoscopic biopsy or surgical specimens, and next-generation sequencing was performed either in-house with a Solid Tumor Panel or with FoundationOne CDx. Specific emphasis was placed on the characterization of double-strand DNA break (DSB) repair genes, as this is the type of tumor cell damage traditionally induced by radiation therapy. Genes associated with the two main pathways of DSB repair, non-homologous end joining (NHEJ) and homologous repair (HR), were analyzed. Specific HR pathway mutations assessed were BLM, BRCA1/2, MRE11, NBN, PALB2, RAD50, RAD51B-D, and RAD54L, while PRKDC mutations were assessed for the NHEJ pathway. Mutations in ATM, an important initiator of DDR pathways, were also analyzed. Additionally, the frequency of mutations in TP53, CDKN2A and SMAD4 in patients with concomitant KRAS mutations was assessed.

RESULTS

Eighty-five patients were included in the study. Five (5.9%) patients had mutations in the NHEJ pathway of the PRKDC gene. Twenty (23.5%) patients had mutations in the HR pathway, including BRCA2 (10/85; 11.8%), PALB2 (5/85; 5.9%), BRCA1 (3/85; 3.5%), and RAD50 (1/85; 1.2%). Six (7.1%) patients had mutations in ATM. No patients were found to have mutations in BLM, RAD51B-D, RAD54L, or NBN. Amongst patients with KRAS mutations (72/85), concomitant mutations were observed in TP53 (47/85; 55.3%), CDKN2A (16/85; 18.8%), and SMAD4 (9/85; 10.6%).

CONCLUSION

Herein, we characterized the frequency of somatic mutations associated with DSB repair genes in patients with BRPC/LAPC. Data analysis on outcomes related to radiation response in patients with mutations in DDR pathways is ongoing, but will likely also benefit from multi-institutional efforts to increase the power to answer this question.

Utilization of the Triangle Volume in Patients with Localized PDAC Undergoing Pre-Operative SBRT: Report of Early Outcomes

PURPOSE/OBJECTIVE(S)

In patients with borderline resectable or locally advanced pancreatic adenocarcinoma (BRPC/LAPC), advances in neoadjuvant therapy have led to an increased proportion of patients undergoing margin negative resection. Nevertheless, locoregional recurrence rates remain high. We have previously reported that the location of locoregional recurrences in this setting map to the "Triangle Volume (TV)," the anatomical space between the celiac artery, superior mesenteric artery, common hepatic artery, and portal vein, which is enriched in extrapancreatic perineural tracts at risk for microscopic residual disease after resection. At the beginning of 2021, we systematically changed our target volume to include the TV, in addition to gross disease and involved vasculature. Herein, we report early locoregional failure outcomes after resection in the setting of BRPC or LAPC treated with pre-operative stereotactic body radiation therapy (SBRT) to the TV, as compared to historical rates.

MATERIALS/METHODS

Patients who received a diagnosis of BRPC or LAPC and who were treated at our institution with neoadjuvant chemotherapy (CTX) and SBRT between 2016 and 2022 were retrospectively reviewed. Between 2016 and 2020, the SBRT clinical tumor volume (CTV) included gross disease and full circumference of involved vasculature at the level of involvement. From 2021 onward, the CTV also included the TV. Survival was estimated using the Kaplan-Meier method. Statistical analyses were performed using scientific 2-D graphing and statistics software.

RESULTS

From January 2016 to December 2022, 204 patients with localized PDAC underwent neoadjuvant CTX followed by SBRT. After completion of SBRT, all patients proceeded with surgical exploration. Of these patients, 111 (54%) had LAPC and 92 (45%) had BRPC disease. All patients were treated with induction CTX, mostly commonly with FOLFIRINOX (N = 166, 81%). Following CTX, the most frequently used SBRT regimen was 33 Gy in 5 fractions (N = 191, 94%). 155 (67%) patients were treated between 2016 and 2020 to the traditional CTV, while 49 (24%) patients were treated after 2020 to a CTV that included the TV. The 2-year local progression free survival rate of patients treated with SBRT using the TV was 77.6% as compared to 47.5% in patients treated with the traditional CTV. Over a median follow up of 15.7 months (range: 1 to 78.2 months), 47% (N = 73 out of 155) of patients who underwent SBRT with the traditional CTV developed locoregional recurrence, but only 12% (N = 6 out of 49) treated with SBRT to the TV have thus far developed locoregional recurrence (p<0.0001).

CONCLUSION

In patients with localized PDAC who undergo pre-operative SBRT for BRPC/LAPC, targeting the TV may help reduce locoregional recurrence. More data and longer follow-up are needed to verify these findings and inform whether the TV may serve as a new standard for target volume delineation in this setting.

Local Time-to-Event Endpoint Under-Reporting and Variability in Pancreatic Cancer Trials Involving Radiotherapy

PURPOSE/OBJECTIVE(S)

The role of radiotherapy (RT) for pancreatic adenocarcinoma (PDAC) remains controversial, with recent studies showing conflicting results. Importantly, endpoints used to evaluate efficacy in recent RT trials for PDAC have been highly variable. As variability in time-to-event (TTE) endpoint definitions is demonstrated to influence outcomes in other cancers, it is critical that radiation oncologists develop consensus around optimal endpoint definitions to use in future PDAC trial design. Thus, we conducted a systematic review of PDAC trials involving RT to characterize the frequency and variability in local TTE endpoint reporting.

MATERIALS/METHODS

An electronic database search was conducted of PubMed, EMBASE, and Cochrane Library to identify phase 2 and 3 clinical trials published from 2010-2022 of localized PDAC involving RT that reported any TTE endpoint (e.g., local control). After excluding duplicates, two independent reviewers screened full-text manuscripts for inclusion. Trial characteristics and local TTE endpoints/definitions were tabulated.

RESULTS

Three hundred twenty references were screened and 79 trials were included, of which 73 (92%) were phase 2 and 26 (33%) were randomized. Twenty (25%) trials reported a local TTE endpoint; these were local control (LC; N = 6), local progression-free survival (LPFS; N = 4), freedom from local progression (N = 6), locoregional progression-free interval (N = 1), cumulative incidence of local recurrence (N = 1), time to failure of sustained LC (N = 1), and local disease-free survival (N = 1). LC (N = 6) had 5 unique definitions and was undefined once; 1 definition included death as an event. LPFS (N = 4) had 3 definitions; 2 did not consider death an event. Among trials with local TTE endpoints, 9 trials specified the definition of a local recurrence/progression. Four trials defined local recurrence based on RT volumes; one counted clinical evidence of recurrence (e.g., tumor bleed); and one counted a rise in tumor markers without evidence of distant metastases. The index time ("time-zero") was defined for local TTE endpoints in 10 trials, including start of RT (N = 4) or chemo (N = 1), end of RT (N = 1), diagnosis (N = 1), enrollment (N = 1), and time of surgery (N = 1).

CONCLUSION

Few pancreatic cancer trials involving RT report local TTE endpoints, with significant heterogeneity in endpoints used and their definitions. Development of consensus endpoint definitions will be critical for future PDAC trial design.

Visceral Artery Pseudoaneurysm Rates after Pancreatoduodenectomy in Patients Who Received Pre-Operative Radiation for Pancreatic Cancer

PURPOSE/OBJECTIVE(S)

The role of pre-operative radiation therapy (RT) for localized pancreatic ductal adenocarcinoma (PDAC) with peri-pancreatic vascular involvement remains controversial, with two recent randomized controlled trials, namely PREOPANC-1 and Alliance A021501, showing conflicting results. Opponents of pre-operative radiation point towards the unclear oncologic benefit and potential added toxicity. While hemorrhage from a pseudoaneurysm (PsA) rupture is a known rare but potentially fatal complication after pancreaticoduodenectomy (PD), with published incidence rates of around 2 to 5%, it has been increasingly common in the pancreatic cancer surgical community for pre-operative RT to be cited as a risk factor for PsA development, despite the absence of data supporting this notion. Certainly, accurate characterization of relevant RT-related toxicities is critical to prevent inaccurate decision-making regarding foregoing its use. As such, we aim to report on the incidence of PsA in a cohort of patients with PDAC who underwent pre-operative RT prior to PD at a high-volume center.

MATERIALS/METHODS

Consecutive patients treated with pre-operative RT via stereotactic body radiation therapy (SBRT) or intensity modulated radiation therapy (IMRT) prior to PD for borderline resectable or locally advanced PDAC were retrospectively reviewed. Incidence of radiographic or clinically apparent PsA was reported. We also characterized the timing of PsA identification in relation to surgery and RT, the artery in which the PsA developed, and the clinical outcomes of patients after PsA identification.

RESULTS

One hundred seventy-five patients met eligibility criteria for our analysis. Most of our cohort (163 patients, 93%) received SBRT to a median dose of 6.6 Gy x 5 (median BED10 54.78 Gy, range: 48 Gy - 61.92 Gy), and only 12 patients (7%) received IMRT in various fractionation and dose patterns (median BED10 62.94 Gy, range 59.47 Gy - 97.5 Gy). The median time between surgery and last contrast-enhanced abdominal imaging was 17 months (range: 0.23 - 68 months). There were fourteen visceral arteries among thirteen patients (7%) that were found to have a PsA on routine follow-up imaging or after a post-PD hemorrhage. The median time between completion of radiation and surgery to PsA were 19.5 weeks (range: 8.6 to 98.1 weeks) and 13 weeks (range: 1.6 to 87.9 weeks), respectively. The two most commonly involved arteries were the gastroduodenal and superior mesenteric arteries. Rate of PsA development was similar among patients treated with SBRT (7%) and IMRT (8%). In terms of Clavien-Dindo classification of complications, there were six patients with grade 3a complications, five patients with grade 4b complications, and two patients with grade 5 complications.

CONCLUSION

Compared to historical data, pre-operative RT does not appear to significantly increase the risk of PsA development after PD. More data on the impact of pre-operative radiation dose-fraction regimen and longer follow-up are needed.

The Prevalence and Management of Synchronous Prostate and Rectal Cancer

PURPOSE/OBJECTIVE(S)

Synchronous prostate and rectal cancer is rare and guidelines for co-management are not well established. This case series explores the prevalence of synchronous diagnosis and different treatment paradigms to propose a standardized approach to management.

MATERIALS/METHODS

We retrospectively reviewed all radiation treatments between 1/2017 and 12/2022 for curative intent treatment to both prostate and rectal cancer. Synchronous was defined as rectal or prostate cancer diagnosed within a 6-month period of each other. We collected baseline characteristics and treatment paradigms including the sequencing of chemoradiation (CRT), chemotherapy (CT), prostate boost, and surgery.

RESULTS

There were 10 out of 2204 total treated patients with prostate or rectal primary noted to have a synchronous diagnosis (0.45%). Table 1 shows characteristics and treatment approach for all patients with 50% receiving CRT and 50% CT alone first. At a median FU of 21.4 months, 2 patients did not complete therapy due to patient choice and both had progression of disease (POD). After completion of CRT, 6 patients underwent rectal surgery with 2 pathological complete response, and 2 patients proceeded with a Watch and Wait approach with clinical complete response on MRI. Prostate boost was delivered equally as often pre-surgery as post-surgery with both SBRT, EBRT and Seed Implant used. There was no grade 3+ RT related toxicity in the patients who completed all therapy.

CONCLUSION

This series represents one of the largest synchronous prostate and rectal cancer cohorts treated with curative intent. Future collaborative work is needed to develop guidelines in the treatment of synchronous prostate and rectal cancers. Although a rare diagnosis, the heterogeneity of approaches has led us to propose a standardized approach to management of synchronous diagnosis with upfront chemotherapy followed by EBRT inclusive of prostate and rectum followed by boost via brachytherapy (SBRT in non-candidates).

Treatment Terminations during Radiation Therapy: A Ten-Year Experience

PURPOSE/OBJECTIVE(S)

Disruptionsin cancer care adversely affect clinical outcomes, particularly when a patient does not complete the prescribed course of treatment. The impact of treatment termination (TT) during radiation therapy has not been well studied. This study addresses TT in a large multi-center department of radiation oncology over a 10-year time period.

MATERIALS/METHODS

TTs of patients undergoing radiation treatment between January 2013 and December 2022 were prospectively tracked as part of departmentalquality and safety monitoring. A TT was defined as the discontinuation of therapy at any point following consent and simulation. Causes of TTs were categorized as: hospice/death, physician choice related to toxicity, physician choice unrelated to toxicity, patient choice related to toxicity, patient choice unrelated to toxicity, progression of disease, non-cancer illness, or other. The rate of TT was calculated as a percentage of all new patients who start radiation treatments. As part of our ongoing department quality and safety program, incremental changes were made to pre-treatment evaluation and scheduling processes, collectively referred to as the "No-Fly" policy. TT rates during three iterations of this policy were compared.

RESULTS

Outof 28,707 planned treatment courses, a total of 1,467 TTs were identified (5.1%). 688 (46.9%) involved patients treated with curative intent, 770 (52.5%) with palliative intent, and 9 (0.6%) for benign disease. The rate of TT decreased from 9.3% in 2013 to 3.3% in 2022. Relative to evolutions of our No-Fly policy, the overall TT rate decreased from 8.8% under No-Fly 1 (2013-2014), to 5.2% during No-Fly 2 (2015-2018), and 4.0% with No-Fly 3 (2019-2022) (ANOVA, p<0.001). The most common sites for TT were H&N (19.3%), CNS (17.9%), and Bone Metastases (17.9%). The most common cause of TT was hospice and/or death (36.5%), 69.1% of which were in patients receiving palliative treatments. Other common causes included patient choice unrelated to toxicity (35%), physician choice unrelated to toxicity (8.8%), and progression of disease (7.6%). There were 473 TTs without radiation dose given (1.6% of planned treatments, 32.3% of TTs).

CONCLUSION

Radiation TTs reflect major deviations from the original care plan. This large cohort study highlights the value of open departmental discourse about TTs, which prompted quality improvement changes that reduced TTs over time. Future studies addressing clinical outcomes can direct treatment decision-making and improve care for our patients.

The Association of Chemoradiation Induced Lymphopenia with Racial Disparity and Its Prognostic Impact on Survival for Anal Cancer

PURPOSE/OBJECTIVE(S)

While the association between chemoradiation induced lymphopenia (CIL) and poor overall survival (OS) is established in multiple solid malignancies, it has not been studied in anal cancer. Racial and socioeconomic disparities as potential predictors of lymphopenia have not been reported. We hypothesize that race and socioeconomic status is associated with increased incidence of severe CIL, which can predict worse overall survival for patients with anal cancer.

MATERIALS/METHODS

A cohort of 75 patients treated with definitive chemoradiation (CRT) for squamous cell anal cancer from January 2014 to December 2020 was reviewed. Total lymphocyte counts (TLC) at baseline and TLC nadir at 1 month post-CRT were analyzed. Logistic regression was used to identify associations between race, gender, ethnicity, median household income by zip code, marital status, baseline hematopoietic cell counts, and post-CRT Grade 3+ lymphopenia (TLC <0.5k/μL). Kaplan-Meier method and Cox regression model were used to perform survival analysis.

RESULTS

Of the 75 patients identified, mean age was 66.9 years and median follow-up time was 37.1 months. There were 63 females, 53 non-Hispanic whites, 22 minorities (12 Blacks, 9 Hispanics, 1 Asians) Radiation dose ranged from 41.4 Gray to 56 Gray. At 1 month post CRT, 85.3% developed lymphopenia (G1 9.3%, G2 26.7%, G3 37.3%, G4 12.0%). On multivariate logistic regression, non-white race demonstrated a trend to have more Grade 3+ lymphopenia (OR = 3.5, p = 0.07). On univariate Cox regression, poorer overall survival was associated with race (HR 3.7, p = 0.04), baseline white blood count (HR 1.3, p = 0.04), baseline hemoglobin (HR 0.6, p = 0.04), and post-CRT Grade 3+ lymphopenia (HR 5.8, p = 0.03). On multivariate Cox regression, only post-CRT Grade 3+ lymphopenia was associated with worse OS (HR 7.5, p = 0.049). 5-year OS significantly differed between patients with and without post-CRT Grade 3+ lymphopenia (62.3% vs 94.7%, P = 0.01).

CONCLUSION

Lymphopenia is commonly observed after chemoradiation for anal cancer. Racial disparity is associated with severe lymphopenia induced by chemoradiation, which is a robust predictor of poor survival in anal cancer. More attention to lymphopenia induced by chemoradiation for anal cancer is needed, particularly in racial minorities.

Efficacy and Safety of Reirradiation with Stereotactic Body Radiation Therapy for Locally Recurrent Pancreatic Adenocarcinoma

AIMS

The purpose of this study was to report on outcomes of a cohort of patients who were treated with reirradiation with stereotactic body radiation therapy (SBRT) for locally recurrent pancreatic adenocarcinoma.

MATERIALS AND METHODS

Patients treated with SBRT reirradiation for locally recurrent pancreatic adenocarcinoma from December 2009 to April 2020 were included in the study. Descriptive statistics were used to record patient demographics, tumour and treatment characteristics. Kaplan-Meier analysis was used to evaluate overall survival, local progression-free survival (LPFS), distant metastasis-free survival and progression-free survival (PFS).

RESULTS

In total, 27 patients were included in the study. The median follow-up time from local recurrence was 19.7 months (range 4.2-43.1 months). Most patients received five-fraction SBRT (26/27, 96%). The median overall survival after local recurrence treatment was 18.3 months (range 3.0-42.6 months), with 6-month, 1-year and 2-year overall survival rates of 88.5%, 73.1% and 33.6%. The median LPFS after local recurrence treatment was 16.2 months (range 2.3-33.6 months), with 6-month, 1-year and 2-year LPFS rates of 95.8%, 62.9% and 27.2%. Peri-SBRT chemotherapy improved LPFS (median 17.5 versus 8.5 months; P = 0.010) and overall survival (median 19.3 versus 5.5 months; P = 0.049). Tumours ≤ 3 cm in the greatest dimension showed better local control (median LPFS 19.2 versus 10.2 months; P = 0.130). There was one case (4%) of acute grade 3 pain and one case (4%) of late grade 3 gastrointestinal toxicity.

CONCLUSIONS

Reirradiation with five-fraction SBRT is safe, but local control remains suboptimal. Patients with smaller tumours experienced improved outcomes, as did patients whose treatment plan included the administration of peri-SBRT chemotherapy.

Impact of a single-day multidisciplinary clinic on the management of patients with liver tumours

PURPOSE

Multidisciplinary cancer clinics may improve patient care. We examined how a single-day multidisciplinary liver clinic (mdlc) affected care recommendations for patients compared with the recommendations provided before presentation to the mdlc.

METHODS

We analyzed the demographic and clinicopathologic data of 343 patients assessed in the Johns Hopkins Liver Tumor Center from 2009 to 2012, comparing imaging and pathology interpretation, diagnosis, and management plan between the outside provider (osp) and the mdlc.

RESULTS

Most patients were white (n = 259, 76%); median age was 60 years; and 146 were women (43%). Outside providers referred 182 patients (53%); the rest were self-referred. Patients travelled median of 83.4 miles (interquartile range: 42.7-247 miles). Most had already undergone imaging (n = 338, 99%) and biopsy (n = 194, 57%) at the osp, and a formal management plan had been formulated for about half (n = 168, 49%). Alterations in the interpretation of imaging occurred for 49 patients (18%) and of biopsy for 14 patients (10%). Referral to the mdlc resulted in a change of diagnosis in 26 patients (8%), of management plan in 70 patients (42%), and of tumour resectability in 7 patients (5%). Roughly half the patients (n = 174, 51%) returned for a follow-up, and 154 of the returnees (89%) received treatment, primarily intraarterial therapy (n = 88, 57%), systemic chemotherapy (n = 60, 39%), or liver resection (n = 32, 21%). Enrollment in a clinical trial was proposed to 34 patients (10%), and 21 of the 34 (62%) were accrued.

CONCLUSIONS

Patient assessment by our multidisciplinary liver clinic had a significant impact on management, resulting in alterations to imaging and pathology interpretation, diagnosis, and management plan. The mdlc is an effective and convenient means of delivering expert opinion about the diagnosis and management of liver tumours.

Tissue biomarkers for prostate cancer radiation therapy

Prostate cancer is the most common cancer and second leading cause of cancer deaths among men in the United States. Most men have localized disease diagnosed following an elevated serum prostate specific antigen test for cancer screening purposes. Standard treatment options consist of surgery or definitive radiation therapy directed by clinical factors that are organized into risk stratification groups. Current clinical risk stratification systems are still insufficient to differentiate lethal from indolent disease. Similarly, a subset of men in poor risk groups need to be identified for more aggressive treatment and enrollment into clinical trials. Furthermore, these clinical tools are very limited in revealing information about the biologic pathways driving these different disease phenotypes and do not offer insights for novel treatments which are needed in men with poor-risk disease. We believe molecular biomarkers may serve to bridge these inadequacies of traditional clinical factors opening the door for personalized treatment approaches that would allow tailoring of treatment options to maximize therapeutic outcome. We review the current state of prognostic and predictive tissue-based molecular biomarkers which can be used to direct localized prostate cancer treatment decisions, specifically those implicated with definitive and salvage radiation therapy.

The potential of shape-based treatment plan optimization for pancreatic IMRT treatments to spare organs at risk and allow for dose escalation to the tumor PTV

316

Background: Due to the low dose tolerance of the organs at risk (OARs) in the abdomen the tumor dose for pancreatic cancer patient is restricted to 50-60 Gy in 1.8-2.0 Gy fractions when combined with chemotherapy. The goal of this study was to develop a system that can determine the minimal radiation dose to the OARs of each individual patient that is achievable while maintaining adequate tumor coverage. This could guide treatment planners to spare the OARs to the fullest extent. When the minimal doses to the OAR are achieved, the total plan can be upscaled until the normal tissue dose constraints are met, allowing for an increase in tumor dose without increased normal tissue toxicity.

Methods: The minimal achievable dose to the OARs depends on its proximity to the planning target volume (PTV). The overlap volume histogram (OVH) was used to describe the spatial relation of each OAR to the PTV. A database of 33 patients, treated with IMRT, was queried to find the lowest achieved dose to an organ for any of the prior patients with less favorable PTV-OAR configurations than the current patient. This minimal dose must also be achievable for the OAR of the new patient. For 25 randomly chosen patients the lowest achievable dose to the liver and kidneys was predicted this way. Then the patients were replanned to verify if this dose could be achieved. The new plans were compared to the original clinical plans.

Results: After replanning the predicted achievable dose to the liver was realized within 1 and 2 Gy for more than 86% and 96% of the patients respectively. For the kidneys these numbers were 83% and 96%. The average improvement in terms of mean dose was 1.4 Gy (range 0 – 4.6 Gy) for the liver and 1.7 Gy (range 0 – 6.3 Gy) for the kidneys. This would have allowed an increase in PTV dose of on average 5 Gy (range 0-13 Gy) based on the liver and 8.5 Gy (range 0-38 Gy) based on the kidneys compared to the original plan, without an increase in dose to the bowel, cord, and stomach.

Conclusions: The lowest achievable dose to the OARs could accurately be predicted for pancreatic cancer patients within seconds. This can guide dosimetrists to spare the OARs or increase the PTV dose by 5 Gy without increased toxicity.

[Table: see text]

Resectable pancreatic small cell carcinoma: The experience of two institutions and review of the literature

333

Background: Primary pancreatic small cell carcinoma (SCC) is rare, with just over 30 cases reported in the literature. Only 7 of these patients underwent surgical resection with a median survival of 6 months. Prognosis of SCC is therefore considered to be poor, and the role of adjuvant therapy is uncertain. Here we report two institutions' experience with resectable pancreatic SCC.

Methods: Six patients with pancreatic SCC at the Johns Hopkins Hospital (4 patients) and the Mayo Clinic (2 patients) were identified from prospectively collected pancreatic cancer databases and re-reviewed by pathology. All six patients underwent a pancreaticoduodenectomy. Clinicopathologic data was analyzed, and the literature on pancreatic SCC was reviewed.

Results: Median age at diagnosis was 50 years (range 27-60). Half of the patients were male, and half were known smokers. All six masses were limited to the pancreatic head. Median tumor size was 3 cm, and all cases had positive lymph nodes except for one patient who only had five nodes sampled. There was no perioperative mortality, although three patients had postoperative complications. All six patients received adjuvant chemotherapy therapy, five of whom were given cisplatin and etoposide. Of these five patients, three were known to have received radiation, while the remaining two had a plan for radiation at an outside facility. Median survival was 20 months with a range of 9-173 months. The patient who lived for 9 months received chemotherapy only, while the patient who lived for 173 months was given chemoradiation with cisplatin and etoposide and represents the longest reported survival time from pancreatic SCC to date.

Conclusions: Pancreatic SCC is an extremely rare form of cancer with a poor prognosis. Patients in this surgical series showed improved survival rates when compared to prior experiences with both resected and unresectable cases. Cisplatin and etoposide appears to be the preferred chemotherapy regimen, although its efficacy remains uncertain, as does the role of combined modality treatment with radiation.

No significant financial relationships to disclose.

Bioluminescence image-guided irradiation and tumor monitoring in a preclinical pancreatic cancer mouse model

192

Background: Preclinical pancreatic cancer animal models for radiation research are far from optimal because they utilize nonlocalized, single-beam irradiation of large fields due to lack of accurate targeting and delivery. We report on a novel preclinical pancreatic cancer research model that utilizes bioluminescence imaging (BLI)-guided irradiation (RT) of orthotopic xenograft tumors, sparing of surrounding normal tissues and quantitative, noninvasive longitudinal assessment of treatment response.

Methods: In accordance with institutional guidelines, luciferase-expressing MiaPaCa-2 pancreatic carcinoma cells were used to generate orthotopic pancreatic tumors in nude mice. BLI of tumors were correlated to PET/CT and necropsy specimens using Pearson correlation. BLI was compared to cone-beam CT (CBCT) to determine the location of the tumor centroid and estimate an appropriate margin for radiation planning. Off-line fusion of BLI with CBCT was performed to guide radiation delivery to tumors using our small animal radiation research platform (SARRP). RT-induced DNA damage was assessed by γ-H2Ax and p-ATM foci. BLI was used to longitudinally monitor radiation treatment response and was correlated to necropsy specimen.

Results: BLI accurately predicted tumor volume (R2 = 0.9961) and correlated well with PET/CT imaging of tumors (R2 = 0.97). BLI centroid accuracy was 3.5 mm relative to that of the CBCT. Irradiated pancreatic tumors stained positively for γ-H2Ax and p-ATM, while surrounding organs were spared. Longitudinal assessment of irradiated (5 Gy) tumors with BLI revealed a significant tumor growth delay of 20 days relative to untreated controls. This was also confirmed pathologically as mean tumor volume of irradiated mice was 30.2% that of unirradiated mice (p < 0.05).

Conclusions: We have developed a bioluminescent, orthotopic preclinical pancreas cancer model that allows noninvasive 1) normalizing of pretreatment tumor burden; 2) treatment planning and image-guided focal RT therapy; and 3) longitudinal assessment of treatment response. This unique translational model offers a means to investigate targeted and systemic agents with focused RT for pancreatic cancer.

No significant financial relationships to disclose.

Prospective assessment of symptoms and quality of life in localized rectal cancer patients receiving chemoradiotherapy

504

Background: Neoadjuvant conformal chemoradiotherapy (CRT) is an important component of treatment for locally advanced rectal cancer, yet its morbidity has not been well characterized using quality of life (QOL) instruments. The present study attempts to establish a baseline distribution of QOL scores before, during, and after CRT and to correlate these changes with symptoms.

Methods: Patients undergoing 3-4 field neoadjuvant CRT for localized rectal cancer were prospectively enrolled at two institutions. Fifty patients completed the QOL instruments at three time points: pretreatment, week 4 of treatment, and 1 month post-treatment. QOL information was captured using three validated questionnaires, the EORTC QLQ-30, QLQ-38, and QLQ-29. Additionally, institutional symptom inventories and CTCAE toxicity data were collected.

Results: Average age was 59.2 years and 72% were men. During CRT, patients had a statistically significant decline in global QOL (70 to 60, p = 0.0024), which normalized (71) following completion of treatment. During therapy, patients also experienced a significant increase in GI symptoms (21 to 27, p = 0.028), urinary symptoms (16 to 30, p < 0.0001), male sexual dysfunction (23 to 34, p = 0.013), and chemotherapy related side effects (8 to 20, p = 0.0001). While these measures returned to baseline 1 month post-CRT, overall sexual function (25 vs. 37, p = 0.0062) and sexual enjoyment (53 vs. 67, p = 0.0070) remained persistently low compared to pretreatment levels. Diarrhea (27%) and proctitis (22%) were the most common grade 3 toxicities. Those patients who experienced grade 3 toxicity during treatment showed markedly decreased global QOL (mean difference = 34).

Conclusions: While rectal cancer patients experienced impaired QOL during neoadjuvant CRT, the vast majority of measures normalized one month after treatment. In contrast, significantly decreased sexual function and enjoyment persisted. This data can be used as a baseline to compare future neoadjuvant conformal CRT regimens and/or assess the toxicity and QOL of new RT modalities such as intensity modulated radiation therapy.

No significant financial relationships to disclose.

Correlation between pancreatic tumor size as measured on 3D CT scan versus pathologic specimen: Impact on radiation treatment volume

276

Background: Definition of the target volume for irradiation of pancreatic cancer (PCA) must balance coverage of micrometastatic disease with toxicity. To determine differences between radiographically defined tumors and true pathologic tumor specimens, we correlated the maximum tumor diameter (TD) of preoperatively imaged tumors with resected tumor specimens.

Methods: With IRB approval, a retrospective chart review of patients who underwent resection of PCA between 2006 and 2010 was conducted. 73 patients were identified with preoperative CT imaging and pathologic analysis of tumors. 70 of 73 patients had a preoperative 3D CT performed. The TD as measured by a radiologist (EF) on contrast CT and 3D CT reconstruction was compared with that measured by pathological analysis of the resected specimen.

Results: 70 patients underwent resection with preoperative CT imaging; 14.1% of these patients had CT performed >6 weeks prior to surgery. The mean (SD) pathologic maximum TD was 31.3 mm (11.3) with range 3 mm to 60 mm. Whereas TD was underestimated by 1.9 mm (1.7 SE) with CT relative to pathologic analysis, this difference was not statistically significant (paired t-test, p=0.27) with a correlation coefficient of 0.265. 3D CT imaging had a smaller mean difference with a mean 3D CT diameter 0.4 mm (1.76 SE) larger than the pathologic specimen (p=0.82) with correlation coefficient 0.222. However, the max TD on 3D CT imaging was on average 2.3 mm larger than on CT (p=0.016) with correlation coefficient 0.798. Of patients with R0 resections (N=48), CT underestimated path size by 3.1 mm (p=0.020), whereas 3D CT was slightly larger (0.1 mm, p=0.949). For R1 resections (n=22), both CT and 3D CT overestimated size (0.8 mm and 1.1 mm, respectively, p>0.5).

Conclusions: PCA TD is generally underestimated on CT imaging, yet better approximated with 3D CT. Improved correlation was seen between CT and pathologic specimens following R0 resection. Alternatively, R1 resection specimens were slightly overestimated by CT/3D CT imaging. As a result, clinical target volumes should be expanded accordingly during radiotherapy planning to properly account for these discrepancies in the gross tumor.

No significant financial relationships to disclose.

Effect of inhibition of poly-(ADP ribose) polymerase on gemcitabine and radiation-induced cytotoxicity of pancreatic cancer cells

203

Background: Poly-(ADP ribose) polymerases (PARPs) are DNA-binding proteins involved in DNA repair. PARP inhibition has resulted in excellent antitumor activity when used with other cytotoxic therapies. ABT-888 is a promising PARP inhibitor with excellent potency against the PARP-1/2 enzymes and good oral bioavailability. We attempt to determine whether PARP-1/2 inhibition alone, or in combination with gemcitabine, will enhance the effects of irradiation (RT) of pancreatic cancer cells.

Methods: The pancreatic carcinoma cell lines, MiaPaCa-2 and Panc02, were treated with ABT-888, gemcitabine, RT, or combinations thereof. RT was delivered with a 137-Cs Gammacell in a single fraction. Cells were pre-treated once with ABT-888 and/or gemcitabine 30 minutes prior to RT. Viability was assessed through reduction of resazurin into fluorescent resorufin. Levels of apoptosis were determined by measuring caspase-3/7 activity using a luminescent assay. PARP activity was determined using a chemiluminescent PAR elisa.

Results: The half maximal inhibitory concentration (IC50) of RT was 5 Gy; IC10 for ABT-888 and gemcitabine were 10 uM and 5 nM, respectively. Treatment with ABT-888 (10 uM), gemcitabine (5 nM), or combinations of the two with RT led to increasingly higher rates of cell death 8 days after treatment (p<0.001). RT dose enhancement factors were 1.5, 1.82 and 2.36 for 1, 10 and 100 uM ABT-888, respectively. Minimal cytotoxicity was noted when cells were treated with ABT-888 alone up to 100 uM. Caspase activity was not significantly increased when treated with ABT-888 (10 uM) alone (1.28 fold, p=0.077), but became significant when RT (2 Gy) was added (2.03 fold, p=0.006). This difference was further enhanced by the addition of gemcitabine (2.95 fold, p=0.004).

Conclusions: ABT-888 is a potent radiosensitizer of pancreatic cancer cells with minimal cytotoxicity when used alone. Cell death is further potentiated by cotreatment with gemcitabine. Radiation-induced apoptosis was significantly enhanced by ABT-888 and gemcitabine, suggesting a synergistic mechanism of interference with DNA repair. These data are currently being validated in an orthotopic pancreatic cancer mouse model.

No significant financial relationships to disclose.

Phase II trial of neoadjuvant full-dose gemcitabine, oxaliplatin, and radiation (RT) in patients with resectable (R) or borderline resectable (BR) pancreatic cancer

239

Background: We previously developed a regimen of full dose gemcitabine (gem), oxaliplatin (ox) and RT to maximize systemic and loco-regional disease control in pancreas cancer (JCO 25:4587, 2007). A multi-institution phase II trial was conducted to test efficacy of this regimen as neoadjuvant therapy in R/BR disease.

Methods: Eligibility criteria included path confirmation, no metastasis, R/BR lesion, PS ≤2, and adequate organ function. Treatment consisted of 2, 28 day cycles of gem (1g/m2 over 30 min D1, 8, 15) and ox (85mg/m2 D1, 15) with RT during cycle 1 (30Gy in 2Gy fractions). Pts were then re-evaluated for surgery. Resected pts received 2 additional cycles of chemotherapy.

Results: 68 evaluable pts were treated at 4 centers in 2007-2010. Median age was 64 (42-83), 32 men, PS 0:1:2 in 40:26:2. Median tumor size 3.2 cm (1.4–7.8), lesion in head 49, body 9, tail 10, R in 24 and BR in 44. 66 pts (97%) completed cycle 1/RT and 61 (90%) cycle 2. Therapy related adverse events ≥ grade 3 in cycles1/2 included ANC (32%), plts (24%), GI (16%), biliary/cholangitis (15%). Best response in primary was partial (10%) or stable (81%). 20 pts not operated on protocol: 8 progression (4 local, 4 distant), 8 judged not resectable, 3 toxicity, 1 early death. Of 48 laparotomies,10 not resected due to vascular involvement (6) or M1 disease (4). Resection completed in 15 of 18 R pts (13 R0, 2 R1) and 23 of 30 BR pts (19 R0, 2 R1, 2 R2). 26 pts received post-op therapy. With median fu 11.3 mos (0.7-35), 42 pts are alive, 20 pts are NED. Median survival (OS) for all pts is 21.2 mos (95%CI 13.3-not defined [ND]), resected 31.1 mos (95%CI 13.7-ND), unresected 16.0 mos (95%CI 5.8-ND). Time to treatment failure (death, progression, toxicity, no resection) and OS in R pts are 9.1 mos (95%CI 2.4-23.8) and 31.1 mos (95%CI 9.8-ND) and in BR pts 5.5 mos (95%CI 2.4-11.8) and 18.0 mos (95%CI 13.3-ND). Correlation of pathologic response and outcome is ongoing.

Conclusions: Neoadjuvant therapy with full dose gem, ox and RT was possible in a large proportion of pts with localized pancreas cancer and resulted in a high percentage of R0 resections. Results are particularly encouraging given a majority of pts with BR disease.

[Table: see text]

Effect of diferuloyl methane (curcumin) on radiation-induced inhibition of proliferation and cytotoxicity of pancreatic cancer cells

222

Background: Pancreatic ductal adenocarcinoma carries a dismal prognosis. Unfortunately, the majority of patients present with locally advanced or metastatic disease unamenable to resection, and the benefit of radiation therapy (RT) in this population is not well substantiated. Diferuloyl methane (curcumin) is a naturally occurring polyphenol derived from the spice turmeric. Although it has been shown to have anticancer properties, it has not been well studied with radiation, particularly for pancreatic cancer. Herein, we attempt to determine whether curcumin can sensitize pancreatic cancer cells to radiation.

Methods: The pancreatic carcinoma cell lines, MiaPaCa-2 and Panc02, were utilized for studies. Cells were treated with curcumin (10 nM to 1 mM), RT (1 to 9 Gy), or combinations thereof and compared to untreated controls. Cells were irradiated using a 137-Cs GammaCell in a single fraction. If applicable, cells were pretreated with curcumin one hour prior to RT. Cell viability was assessed through reduction of resazurin into fluorescent resorufin. Levels of apoptosis were determined by measuring caspase-3 and -7 activities using a luminescent assay.

Results: Treatment with curcumin alone led to minimal inhibition of proliferation until concentrations exceeded 10 uM. The half maximal inhibitory concentration (IC50) was 40 uM and the IC20 was 10 uM. Cells experienced a dose- dependent inhibition of proliferation with increasing RT doses (IC20 of 5 Gy). Combined treatment with IC20 of curcumin and RT led to a synergistic increase in inhibition of proliferation (70%). The RT dose enhancement factor was 2.5 with 100 uM curcumin and 5 Gy. Caspase 3/7 activity was not significantly enhanced by treatment with curcumin alone (10 uM), but increased significantly when RT was added (5 Gy; p<.01).

Conclusions: Curcumin is a potent radiosensitizer of pancreatic cancer cells that synergistically enhances the effects of RT. It has minimal effects on proliferation and apoptosis when used as a single agent at lower doses. Curcumin also potentiates RT-induced cytotoxicity through induction of apoptosis. These data are currently being validated in an orthotopic pancreatic cancer model.

No significant financial relationships to disclose.

Multicenter evaluation of adjuvant therapy for gallbladder cancer

251

Background: To assess the effect of adjuvant therapy in gallbladder adenocarcinoma (GBC).

Methods: Retrospective review was conducted at five institutions to identify pts who had surgery for confirmed dx of GBC from 1985-2008 (N = 189). Pts were excluded if they had chemo alone (N = 8), path other than adenoca (N= 7), carcinoma in situ (N=1), < 30 days of follow-up (N = 2), or missing data (N=14). Of the remaining 156 pts, 58 received surgery only and 98 received adj RT ± chemo. Kaplan-Meier was used for overall survival (OS) and Cox proportional hazards to compare risk factors.

Results: Median age of dx was 64.4, 68.0% were female, 37.9% had ≥ stage 2b, 37.2% had + nodes, and 32.1% had + margins. Overall, 35.9% of the patients had simple cholecystectomy (SC) only and 64.1% had radical resection (ER). mOS for pts treated with surgery alone was 49.7 months (95% CI: 24.8 to Inf). On univariate analysis, + margins (HR 2.72, p<0.001) was associated with worse OS, whereas ER compared to SC improved survival in both univariate (HR 0.46, p<0.001) and multivariate (HR 0.53, p=0.033) analyses after adjusting for node/margins, T-stage, adj RT, age, gender, and institution. mOS for the entire cohort vs. adj RT (median 50.4 Gy) ± chemo was 30.7 months (95% CI: 19.2 to 46.9) vs. 26.9 months (95% CI: 15.5 to 39.1). But, compared to surgery alone, the adj group was more likely to have had node +, margin +, or T-stage 3+ (all p<0.001). The adj RT group was also less likely than surgery alone pts to have undergone ER (p = 0.007). On multivariate analysis, decreased OS was also found for node + (HR 2.09, p=0.004), margin + (HR 1.84, p=0.043), and T3/T4 disease (HR 2.37, p=0.002). After adjusting for surgical extent, node, margin, T stage, age, gender, and institution, there was improved OS with adj therapy (HR: 0.43, p = 0.020). When stratified by surgical extent, the risk estimate for adj RT improved OS among those with SC (n=56; HR 0.20, p=0.135) and ER (n=100; HR 0.46, p=0.067), but was not statistically significant.

Conclusions: ER was associated with improved OS, whereas node/margin+ and T-stage 3+ were associated with worse survival. In multivariate analysis, adj RT improved OS after surgery. Given the poor prognosis of GBC patients with advanced disease, consideration of adj therapy is appropriate.

No significant financial relationships to disclose.

Stereotactic radiation treatment planning with volumetric modulated arc therapy: Impact of duodenal sparing on pancreatic tumor coverage

301

Background: Volumetric modulated arc therapy (VMAT) allows for intensity-modulated radiation delivery with faster treatment times and fewer delivered monitor units (MU). The dose-limiting structure for pancreatic stereotactic body radiation therapy (SBRT) is the duodenum. We evaluate VMAT dose distribution, delivery times, and the effect of duodenal sparing (DS) for pancreas SBRT.

Methods: Plans of 15 patients with unresectable pancreatic cancer (14 head/1 tail) were selected. VMAT treatment planning with the “SmartArc” function of Pinnacle v. 8.9 was used to plan one fraction of 25 Gy to the PTV (gross tumor + 2 mm expansion) normalized to the 80% isodose line. Two VMAT SBRT plans were conducted for each case; the first did not attempt to spare the duodenum (non DS) while the second did (DS). Constraints were stomach/duodenum any point max <30 Gy (for DS plan), liver D50 < 5 Gy, ipsilateral kidney D25 < 5 Gy, cord Dmax < 5 Gy and stomach D4 < 22.5 Gy.

Results: Gross tumor volume ranged from 58.4cm3 to 320.3 cm3. The average overlap volume between PTV and the duodenum was 8.4 cm3. In 10/15 non-DS plans, the duodenal Dmax exceeded 30 Gy. With DS optimization, only 1/15 plans exceeded the 30 Gy threshold. These differences were statistically significant (p<0.001). Typical MU and delivery times, as calculated by the planning software, were 5494 MU and 775 secs vs. 5296 MU and 703 secs for the DS and non-DS plans, respectively. The difference in delivery times was significant (p=0.01), but amounted to only 1.2 min on average. The average duodenal Dmax for non-DS plans was 30.4Gy, D4% was 23.4 Gy. With DS, the average Dmax was reduced to 28.1Gy and D4% to <19.7 Gy (p<0.001). As expected, VMAT plans with greater overlap between the duodenum and PTV had a higher duodenal Dmax.

Conclusions: This study demonstrates the feasibility of VMAT for high-dose SBRT treatment of pancreatic cancer incorporating constraints to limit the dose to the duodenum. Future studies will evaluate whether VMAT with fractionated SBRT results in improved duodenal sparing more efficiently than traditional IMRT.

No significant financial relationships to disclose.

Correlation of DPC4 status with outcomes in pancreatic adenocarcinoma patients receiving adjuvant chemoradiation

168

Background: In an autopsy series of patients with advanced pancreatic cancer (PCA), loss of DPC4 was highly correlated with disseminated metastasis. The purpose of this study was to determine if DPC4 gene status predicts for survival and patterns of recurrence following adjuvant (adj) chemoradiation (CRT).

Methods: 101 patients who underwent surgery followed by adjuvant 5-FU or gemcitabine-based CRT were studied. Imaging studies were reviewed to assess patterns of recurrence and tumor tissue obtained to determine DPC4 immunolabeling status. DPC4 status was graded as intact or lost/mutated. Kaplan-Meier estimates were used to compute survival and Cox proportional hazards were used to compare risk factors.

Results: Median overall (mOS) and progression-free survival (PFS) was 22.6 mos (95% CI 18.8 to 31.6) and 14.0 mos (95% CI 11.5 to 18.4). The mOS and 1-yr OS for patients with DPC4 intact vs. lost status was 21.9 mos (95% CI 16.8-32.4) and 78.4% vs. 22.6 mos (95% CI 18.4-32.6) and 78.2%, respectively (HR: 1.05, p=0.82). After adjusting for node, margin status, tumor grade, tumor size, and age, DPC4 status did not predict for mOS (RR 1.04, 95% CI: 0.62-1.74, p=0.89). Time to first progression at any site for PCA with DPC4 intact vs. lost status was 13.8 vs. 14.0 mos (p=0.79). Local recurrence was more common in PCA with DPC4 loss than with intact status (34.4% vs. 13.7%, p=0.012). There was no difference in the rates of distant recurrence in PCA with intact vs. loss of DPC4 expression (62.8% vs. 55.7%, p=0.45); however, DPC4 loss was more commonly associated with liver recurrence (27.9% vs. 19.6%, p=0.31).

Conclusions: In pancreatic cancer patients receiving adj CRT, loss of DPC4 labeling in their resected PCA indicates a greater likelihood of developing local recurrence despite having received adj CRT. Efforts to improve loco-regional control are therefore needed for these patients following surgery and adj CRT.

No significant financial relationships to disclose.

Adjuvant chemoradiation therapy for pancreatic adenocarcinoma: Impact of family history on outcome

15044

Background: The objective of the current study was to examine the impact of a family history of pancreatic adenocarcinoma (PCA) on the outcome of patients receiving adjuvant chemoradiation therapy (CRT) following pancreaticoduodenectomy (PD). Patients and Methods: Between August 1993 and February 2005, 902 patients underwent PD for pancreatic adenocarcinoma. Following PD, 405 patients received no adjuvant CRT, while 346 patients received CRT. Another 151 patients were excluded because they received protocol treatment, neoadjuvant CRT, or were lost to follow-up. Patients who received adjuvant CRT were treated with 5-FU (97.4%) based CRT (median dose 50 Gy) and maintenance 5-FU or gemcitabine. Survival was estimated using the Kaplan-Meier method and differences in survival were examined using the log-rank test. Cox regression analysis was used to control for family history. Results: Of the 751 patients included in the study, 158 (21%) patients had a known family history of pancreatic adenocarcinoma (only one family member n=119; >=2 either first or second degree relatives, n=39). Clinicopathologic characteristics of patients with a family history of PCA were similar to those of patients who did not have a family history (age, race, positive lymph node status, primary tumor size, and proportion receiving adjuvant CRT; all P>0.05). In an analysis of the entire patient cohort, adjuvant CRT was associated with an improvement in median overall survival compared with no adjuvant CRT (21.0 months vs. 14.6 months, respectively; P= 0.001). Family history of PCA (>=1 family member) was not associated with overall survival (positive family history, 20.0 months vs. negative family history, 17.3 months; P = 0.12). Family history of PCA also did not modify the effect of CRT on overall survival. Specifically, on multivariate analysis, after stratifying on family history (>=1 family member), CRT remained significantly associated with an improved survival (Hazard ratio=0.71; P=0.001). Conclusion: Adjuvant 5-FU based CRT improves the median survival of patients with resected pancreatic adenocarcinoma. The improvement in median survival associated with adjuvant CRT was independent of a familial history of pancreatic adenocarcinoma.

No significant financial relationships to disclose.

The use of carboplatin and paclitaxel with daily radiotherapy in patients with locally advanced squamous cell carcinomas of the head and neck

PURPOSE

Unresectable squamous cell carcinomas of the head and neck (SCCHN) continue to pose a significant therapeutic challenge. This report defines the toxicities, efficacy, and prognostic factors associated with the combination of carboplatin (CBDCA), paclitaxel, and once-daily radiation for patients with locally advanced disease. Additionally, the pharmacokinetics of paclitaxel were investigated.

METHODS AND MATERIALS

From 1993-1998, 62 patients with Stage III-IV SCCHN were treated with 70.2 Gy of RT at 1.8 Gy/fraction/day to the primary site. Weekly chemotherapy was given during RT consisting of paclitaxel (45 mg/m(2)/wk) and CBDCA (100 mg/m(2)/wk). All patients presented with locally advanced disease; 77% had T4 disease and 21% had T3 disease. Fifty-eight percent had N2b-N3 disease.

RESULTS

Sixty patients were evaluable for response and survival with a median follow-up of 30 months (range 7-70). Ninety-eight percent of patients completed prescribed therapy. One patient died after refusing medical management for pseudomembranous colitis and is scored as a Grade 5 toxicity. Two patients suffered Grade 4 leukopenia. Median number of break days was two. A clinical complete response (CR) at the primary site was obtained in 82%, with a total (primary site and neck) CR rate of 75%. The median survival for the entire cohort is 33 months. Response to therapy and status of the neck at presentation were the only prognostic factors found to influence survival. The median survival for patients who attained a CR is 49 months versus 9 months in those who did not attain a CR (p < 0.0001). The 2- and 3-year overall survival for complete responders are 79% and 61%. Plasma paclitaxel concentrations in the range shown to be radiosensitizing were achieved.

CONCLUSIONS

Weekly carboplatin and paclitaxel given concurrently with definitive once-daily external beam radiation therapy is well tolerated with over 90% of patients completing prescribed therapy. An ultimate CR rate of greater than 70% was obtained, which translated directly into improved survival. With 48% 3-year overall survival for the entire group, this regimen is an excellent option for this group of patients with a historically poor prognosis.

An assessment of practice support and continuing medical education needs of rural Pennsylvania family physicians

Family physicians provide the greatest proportion of care in rural communities. Yet, the number of physicians choosing family practice and rural practice has continued to decline. Undesirable aspects of rural practice, such as professional isolation and a lack of or inadequate resources, are assumed to be associated with this decline. This article reports on the practice support and continuing medical education needs of rural family physicians. A mail survey was conducted in 1993 on a purposive sample of family physicians in 39 of 67 rural-designated or urban Pennsylvania counties with low population densities. The physicians identified needs that included patient education materials and programs, community health promotion, federal regulation updates, technical assistance with computers and business management, database software and a videotape lending library, a drug hotline, and mini-fellowships on clinical skill development. A majority of respondents were willing to participate in clinical educational experiences for students and residents. Some physicians indicated a lack of interest in access to information through telecommunications, e.g., video conference referrals and consultations. Overall, findings revealed that family physicians need and are receptive to a variety of practice support and continuing education programs. A practice support program coupled with policy coordination among public and private organizations is likely to lessen complaints by rural primary care physicians.

Factors influencing retention of rural Pennsylvania family physicians

BACKGROUND

Rural areas suffer from a lack of primary care physicians. Efforts to retain physicians should focus on modifying or changing attributes of rural practice that are considered by physicians to be undesirable. A practice support outreach program (PSOP) is one initiative expected to enhance retention in rural areas by addressing negative aspects of rural practice. The purpose of this study was to assess factors related to satisfaction and retention of family physicians to develop and implement a PSOP in rural areas of Pennsylvania.

METHOD

In 1993 a mail survey was conducted on a convenience sample of 398 family physicians practicing in 39 counties in Pennsylvania.

RESULTS

Twenty percent of respondents were considering leaving rural practice. Bivariate analyses indicated that professional isolation, lower reimbursements, and sharing on-call with only 1 other physician were associated with physician's reasons for considering leaving rural practice. A multiple logistic regression revealed that sharing on-call rotation with only 1 other physician and having a solo practice were significant influences in considering leaving rural practice.

CONCLUSIONS

Findings suggest support strategies that minimize perceptions of professional isolation and policy efforts that address reimbursement differentials and compliance issues are needed to minimize many complaints of rural family physicians.

Middle cerebral artery endarterectomy: experience with two cases

BACKGROUND

Atherosclerotic stenosis of the middle cerebral artery (MCA) is uncommon and long-term prognosis is not well defined. Methods of treating stenosis of the MCA range from the administration of anticoagulants to endarterectomy.

METHODS

We present two cases of patients with focal symptomatic stenosis of the MCA with evidence of focally decreased cerebral blood flow and compromise of cerebral blood flow reserves on xenon-enhanced computed tomography (Xe CT) scanning. Endarterectomies were performed after unsuccessful anticoagulation therapy.

RESULTS

Both patients underwent successful endarterectomies of the MCA. Improvement in cerebral blood flow postoperatively was documented for both patients. At last follow-up neither patient had demonstrated any additional ischemic episodes.

CONCLUSIONS

Atherosclerotic stenosis of the MCA may be responsible for distal emboli and compromised hemodynamics, and endarterectomy of this vessel may provide definitive therapy.

Genesis of a dural arteriovenous malformation in a rat model

A rat model was developed to determine the role of sinus thrombosis and elevated sinus pressures in the pathogenesis of dural arteriovenous malformations (AVMs). Five protocols were tested to compare various sinus pressures and thrombosis of a sinus: 1) Control I, sham operation (five animals); 2) Control II, occlusion of the right common carotid artery, the right external jugular vein, and the vein draining the left transverse sinus, as well as thrombosis of the sagittal sinus (10 animals); 3) arteriovenous fistula (AVF) I, anastomosis of the right common carotid artery to the external jugular vein causing retrograde flow through the transverse sinus (10 animals); 4) AVF II, anastomosis (as described in AVF I) and thrombosis of the sagittal sinus (12 animals); 5) AVF III, anastomosis (as described in AVF I) as well as thrombosis of the sagittal sinus and occlusion of the vein draining the transverse sinus on the left (12 animals). Mean arterial and sagittal sinus pressures were monitored and cerebral angiograms were obtained intraoperatively and again 90 days later. Afterward, the animals were sacrificed and their brains and dura were examined histologically. Formation of a fistula resulted in a significant (p < 0.05) threefold increase in sagittal sinus pressure in the AVF II group and a significant (p < 0.05) sixfold increase in the AVF III group. Seven dural AVMs (three in the AVF II group and four in the AVF III group) were demonstrated angiographically and histologically. The seven malformations were located adjacent to a thrombosed sagittal sinus. All lesions were within the dura and sinus wall with direct thrombus-sinus wall connections demonstrated in four of the malformations. The other three lesions displayed arteriovenous connections within the sinus wall and dura. These data suggest the importance of not only sinus thrombosis but also sinus hypertension in the development of a dural AVM.

Ruptured giant aneurysm of an occluded middle cerebral artery in a severe-grade patient: case report

A 27-year-old woman presented with a ruptured giant aneurysm of the right middle cerebral artery, occlusion of the artery distal to the aneurysm, and an intraparenchymal hematoma. Before emergent surgery, her condition deteriorated to a Glasgow Coma Scale score of 4 and a Hunt and Hess grade of V. Electroencephalographic response on the right, initially absent, was present toward the end of the surgery. Postoperative angiography demonstrated good filling of the previously occluded distal middle cerebral artery. The simultaneous occurrence of hemorrhage from a giant intracranial aneurysm and occlusion of the parent artery is extremely rare, and this is the first case we found in the literature in which the patient survived. The treatment strategy and outcome are discussed.

Combined surgical and endovascular treatment of a giant vertebral artery aneurysm in a child. Case report

The case is presented of a 5-year-old child who suffered a subarachnoid hemorrhage from a giant left vertebral artery-posterior inferior cerebellar artery (PICA) aneurysm. Initial treatment consisted of surgical occlusion of the parent vertebral artery combined with a PICA-to-PICA bypass. Because of persistent filling of the aneurysm, the left PICA was occluded at its takeoff from the aneurysm. Endovascular coil occlusion of the aneurysm and the distal left vertebral artery enabled complete elimination of the aneurysm. Follow-up magnetic resonance imaging and arteriography performed 6 months postoperatively showed persistent occlusion and elimination of the mass effect. Combined surgical bypass and endovascular occlusion of the parent artery may be a useful adjunct in the management of these aneurysms.

Cervical corpectomy and plate fixation for postlaminectomy kyphosis

Between 1987 and 1991, 20 patients with symptomatic postlaminectomy kyphosis were treated with anterior decompression, bone graft, and anterior cervical plate. The patients were predominantly male (14:6) with a mean age of 58 years. The initial laminectomy was performed for either spondylosis (80%) or spinal tumor (20%). All patients had anterior compressive pathology, which was associated with instability (45%), neck pain (75%), myeloradiculopathy (90%), or severe neck deformity (30%). The mean degree of kyphosis was 38 degrees. Treatment consisted of a trial of cervical traction (75%), anterior corpectomy (95%), intersegmental decompression (5%), bone fusion (100%), and fixation with either Caspar (85%) or Synthes (15%) anterior plating at a mean of 3.8 levels. Halo fixation was used in 10% of patients. Postoperative complications included vocal cord paresis (15%), pneumonia (10%), wound dehiscence (5%), and screw pull-out (5%). At follow-up evaluation, a mean of 28 months after treatment, all patients had a solid fusion and a mean curvature improvement to 16 degrees residual kyphosis. Neurologically, 10% were cured, 55% were improved and returned to premorbid function, 30% were stable, and 5% had late progression. These data suggest that immediate fixation with anterior plating facilitates solid fusion, maintains spinal curvature, and promotes neurological improvement.

Should all women with cervical atypia be referred for colposcopy: a HARNET study. Harrisburgh Area Research Network

BACKGROUND

Clinicians who manage women with Papanicolaou (Pap) smears showing atypical squamous cells of undetermined significance (ASCUS) may miss clinically significant cervical disease by repeating the cytology alone. We evaluated the ability of the human papillomavirus (HPV) screen and the naked-eye examination after a cervical acetic acid wash to enhance the follow-up Pap smear in predicting an abnormal colposcopic biopsy.

METHODS

Pap smears were performed on all women (N = 7458) attending six family practice offices for a health maintenance examination from August 1989 through February 1991. Consenting subjects with ASCUS underwent repeat cytological testing, an HPV screen, and a cervical acetic acid wash examination immediately before colposcopy after a 4- to 6-month waiting period.

RESULTS

Of the 122 consenting women identified with ASCUS, 67 (55%) demonstrated abnormalities on biopsy, including 26 with condyloma, 26 with cervical intraepithelial neoplasia I (CIN I), and 15 with CIN II to III. The false-negative rate, 58%, of the follow-up Pap smear alone for detecting these cases of condyloma and CIN was significantly decreased (false-negative rate, 27%) with the use of the cervical acetic acid wash as an adjunctive test. There was no additional reduction in the false-negative rate with the use of the HPV screen. Of the 15 subjects with high-grade cervical lesions (CIN II to III), 14 had either an abnormal follow-up Pap smear or an abnormal cervical acetic acid wash examination.

CONCLUSIONS

Among women with cervical atypia, a single follow-up Pap smear alone failed to detect one third of the cases of high-grade disease. Ninety-three percent of these cases were detected, however, with a follow-up Pap smear and an acetic acid wash. Our one subject with a high-grade lesion missed with this combination of tests had an unsatisfactory Pap smear. Use of both tests together may reliably guide clinical decisions regarding the management of cervical atypia.

Cardiopulmonary bypass, hypothermic circulatory arrest and barbiturate cerebral protection for the treatment of giant vertebrobasilar aneurysms in children

Intracranial aneurysms in the pediatric population are rare lesions with a significant proportion occurring as giant aneurysms involving the vertebrobasilar system. The complex anatomy of these lesions frequently creates an extremely difficult management dilemma. We present two patients, a 9-year-old and a 13-year-old, with giant fusiform vertebrobasilar aneurysms, to illustrate the utility of cardiopulmonary bypass, hypothermic circulatory arrest and barbiturate cerebral protection for successful surgical treatment of these complex intracranial vascular lesions in children.

Analysis of 153 patients with myelomeningocele or spinal lipoma reoperated upon for a tethered cord. Presentation, management and outcome

After primary repair of myelomeningoceles or lipomyelomeningoceles, late progressive neurologic deterioration commonly occurs due to a treatable cause. In our experience many of these patients have a tethered cord. With early untethering, most patients are stabilized and a significant percent of the patients show improvement in their clinical status. Of 341 tethered cord releases done from 1981 to 1988, we report on 153 patients reoperated upon following primary repair. One hundred were performed after primary closure of a myelomeningocele and 53 after repair of a lipomyelomeningocele. The average age of the patients with a myelomeningocele was 6 years old, and for the spinal lipoma patients, 8 years old. The presenting symptoms were similar; weakness, deterioration in gait, scoliosis, orthopedic deformities, and urinary incontinence represented the most common complaints. All 153 patients were noted to have a tethered cord at operation. Additional pathology (dermoid tumors, hydromyelia, tight filum and diastematomyelia) was present in 30% of the cases. With the use of the CO2 laser for dissection, all but 10 patients could be untethered. Follow-up over an average of 4 years revealed 93% of the patients with a myelomeningocele had stabilization or improvement of their presenting complaints, and 7% had progression of their presenting complaints. All of the lipomyelomeningocele patients had either stabilization or improvement of their presenting complaints. There were no mortalities. Close follow-up and early treatment of this patient population is indicated. With release of the cord a significant portion of the population will have relief or improvement of their presenting complaints.

Single community research networks. The HARNET experience. Harrisburg Area Research Network

Clinical research performed in family physicians' offices is critical for building an expanded knowledge base for modern health care. Practitioners do not usually have the time, funds, or research expertise to conduct clinical studies. Organized networks can accomplish this goal. Large-area networks, composed of many separate practice sites from a wide geographic area, are valuable sources of information for describing the natural history of disease. These observational studies usually consist of data collection during clinical practice. Experimental trials include evaluations of new protocols, diagnostic tests, or therapies, often in a randomized and blinded fashion. Because of the difficulties in adhering to a standardized protocol, experimental trials are rarely undertaken in the busy clinician's office. Similarly, it may be difficult to standardize these studies in large-area networks. Smaller networks, often in a single community, can feasibly perform more complex studies. Important strategies are required to avoid loss of interest, lack of continuity, and conflict of interest.

Preservation of olfaction in anterior craniofacial approaches

Through the combined efforts of neurosurgeons, head and neck surgeons, and craniofacial surgeons, the standard transbasal approach to the frontal fossa has been modified to include removal of the orbital roofs, nasion, and ethmoid sinuses. This approach has been combined further with facial disassembly procedures to provide extensive midline exposure to the midface and clival region. Extended frontal approaches, however, necessitate removal of the crista galli and sectioning of the olfactory rootlets with the associated risk of anosmia, cerebrospinal fluid (CSF) leak, and the need for complex reconstruction of the frontal floor. To avoid these problems, the authors have modified the technique of handling the cribriform plate to preserve the olfactory unit. Circumferential osteotomy cuts are made around the cribriform plate to allow an en bloc removal with its attachment to both the dura and underlying mucosa. Opening of the dura is avoided and the cribriform bone is used to reconstruct the floor. Four patients underwent this approach, for treatment of an angiofibroma in three and a fibrosarcoma in one. The mean follow-up period was 7 months. No patients developed a CSF leak, and within 8 weeks olfaction had returned in all patients. There was no other associated morbidity. These data suggest that this modification of the transbasilar approach can alleviate extensive reconstructive procedures and CSF leaks while preserving olfaction.

Follow-up Papanicolaou smear for cervical atypia: are we missing significant disease? A HARNET Study

BACKGROUND

The presence of cervical atypia on the Papanicolaou (Pap) smear may be an indicator of significant cervical disease. Many investigators recommend that colposcopy be performed in these women. We wished to determine the prevalence of undetected cervical intraepithelial neoplasia (CIN) among women identified as having cervical atypia by cytologic testing in a primary care setting.

METHODS

Pap smears were performed on all women (N = 7458) attending six family practice offices for a health maintenance examination from August 1989 through February 1991. Cytologic specimens were obtained using an endocervical Cytobrush and wooden spatula. Consenting subjects with cervical atypia underwent repeat Pap smear testing immediately before a colposcopic examination after a 4- to 6-month waiting period.

RESULTS

One hundred fifty-nine women identified as having cervical atypia consented to having a colposcopic examination. Of these, 96 (60%) demonstrated abnormalities on biopsy, including 40 with condyloma, 41 with CIN I, and 15 with CIN II to III. The false-negative rate of the follow-up Pap smear for detecting these cases of condyloma and CIN was 57%.

CONCLUSIONS

One third of the women with cervical atypia identified on an initial Pap smear in this primary care community setting had CIN after colposcopic biopsy. The single follow-up Pap smear obtained with the endocervical Cytobrush and wooden spatula failed to detect one half of the cases of biopsy-proven CIN. Further studies regarding the use of additional screening methods for follow-up are necessary.

Diving accidents. Mechanism of injury and treatment of the patient

Diving accidents that result in cervical spinal injuries most often occur in young adult males who have access to alcohol and water. A C-5 injury with permanent neurologic deficit is the most common pattern of injury in these patients. Recommended treatments include emergent alignment, an intravenous bolus of methylprednisolone, and spinal stabilization. Recovery is a function of the severity of the initial neurologic injury.

Pediatric intracranial aneurysms: simple and complex cases

Between 1984 and 1990, 20 aneurysms in 16 patients (9 males and 7 females), 18 years or younger (mean age, 8 years; range, 7 months to 18 years), were treated at our institution. Seven patients had a solitary saccular aneurysm located at an arterial bifurcation which manifested as a subarachnoid hemorrhage. Six of these patients were treated with clip obliteration. Postoperative results were excellent or good in 5 and fair in 1. Nine patients had complex or multiple aneurysms of variable origins (3 giant, 2 infectious, 2 traumatic and 2 associated with an arteriovenous malformation) and presentation. Surgical treatment of these children required the use of hypothermic arrest, trapping, bypass and anastomotic procedures. Outcome was excellent or good in 7 and fair in 2. An analysis of these patients with regard to pathogenesis and management is presented.

Diving injuries of the cervical spine

From 1975 to 1986, 2435 patients were admitted to the Northwestern University-Midwest Regional Spinal Cord Injury Unit. Of these, 220 patients (9.0%) had documented neck fractures from diving accidents, representing the largest series of acute diving injuries yet analyzed. The average age of these patients was 21 years, and males predominated. Two hundred twelve patients (96.4%) were admitted within 48 hours of injury. Associated injuries were rare: none had intracranial mass lesions or systemic injuries, and only nine were near-drowning victims who required endotracheal intubation. The most common levels of injury were C-5 (140 fractures) and C-6 (85 fractures), with 70 patients having fractures at more than one level. Neurological injury was sustained in 154 (70.0%) patients, while 66 (30.0%) patients were neurologically intact. One hundred forty-seven (66.8%) patients underwent posterior cervical fusion, and anterior fusion was performed in 36 (16.4%), allowing for early ambulation and an average hospital stay of 17 days. Hospitalization was relatively uncomplicated, with urinary tract infection in 121 (55.0%), pneumonia in nine (4.1%), and deep vein thrombosis in 24 (10.9%). Long-term follow up averaged 5 years and was obtained in 160 (72.7%) patients. Sixteen (10.0%) improved neurologically, five (3.1%) deteriorated, and 139 (86.9%) were unchanged. Notably, this large study shows that diving accidents occur in a young, healthy population who sustain essentially no other associated intracranial or systemic injuries and have few serious hospital complications. Such patients may be mobilized early in their care after either internal or external stabilization. Subsequent long-term neurological improvement can be expected to occur in about 10% of patients. The importance of water safety and injury prevention is stressed.

Tethered cord as a cause of scoliosis in children with a myelomeningocele

Scoliosis interferes significantly with the functional ability of most children with a myelomeningocele. While it is recognized that tethered cord at the repair site causes neurological deterioration, it has been controversial whether tethered cord causes scoliosis. The spinal cord was untethered in 30 children with progressive loss of function and scoliosis. Of 6 children with curves greater than 50 degrees only 1 improved. Of the other 24 children their curves were stable or improved at 1 year follow-up. At late follow-up, 2-7 years, 63% were stable or improved while 38% began to progress. Tethered cord causes scoliosis and stability or improvement can be anticipated following untethering. Close long-term follow-up is essential to identify those individuals with retethering of their cord.

Case-control studies

Case-control studies investigate associations between exposures to risk factors and a condition of interest. Subjects with the condition (cases) are compared to subjects known not to have the condition (controls) with respect to the presence of risk factors in the past. This design is particularly useful when the condition of interest is relatively uncommon. Case-control studies are useful for studying the etiology of conditions in primary care. Invalid results may be obtained, however, if a study's design allows the occurrence of selection bias, information bias, or confounding.

Teaching clinical epidemiology: a controlled trial of two methods

The teaching of clinical epidemiology to second-year students at the Bowman Gray School of Medicine is carried out using journal articles to illustrate concepts. Because of the need for discussion, the instructors believed that the concepts of epidemiology might best be learned by, and that greater satisfaction with the learning process might be derived from, small group discussions rather than large lecture sessions. To test these hypotheses, students were randomized into either one of two discussion groups or a larger lecture group. The course handouts and text were identical, and the three instructors presented the same material successively to each group. In the final examination, all three groups answered approximately 26 of 36 questions correctly. Seventy percent of students responded to a questionnaire at the end of the course. There were no significant differences between the discussion and lecture groups in their ability to read and understand medical articles. However, the discussion group students were more favorable in their assessment of the success of the teaching method and in their perception of the importance and overall quality of the course. While there may be little difference in the short-term retention of epidemiological principles between the two teaching methods, the greater satisfaction reported by the students in the small groups will stimulate us to try to provide that type of learning environment in the future.

Utilization review and the family physician

Family physicians are often in close contact with utilization review programs. Traditional hospital based review programs have included prehospital certification, outpatient and same-day surgery, second opinions, concurrent hospital review, and early discharge planning. The role of utilization review in ambulatory care is expected to increase in the future and will focus on the cost-effective allocation of resources. Family physicians need to be aware of several issues as they interact with utilization review programs, including the preservation of confidentiality, financial obligation in the face of negative review decisions, and the effect of utilization review on liability for adverse outcomes. Family medicine education should encourage familiarity with the structure and function of review programs and should enable practitioners to participate intelligently in such programs.

Long-term follow-up of the polytetrafluoroethylene (PTFE) prosthesis as an arteriovenous fistula for haemodialysis

During a 10-year period vascular access in 86 patients receiving long-term haemodialysis was provided by the insertion of 100 polytetrafluoroethylene grafts. In 24 patients (28%) the PTFE prosthesis was used as a primary access operation; 62 patients (72%) had had previous access procedures. Early thrombosis and infection led to graft failure in eight patients. Late fistula occlusion developed in 67 instances in 39 fistulae. Thrombectomies were performed in 53 fistulae with good result in 46 (86.8%). Infection after puncture was seen in nine grafts and was treated by incision and drainage (33%) or graft removal (67%). False aneurysms developed in six patients; prolonged haemorrhage from the puncture site occurred in one patient and haemodynamic complications (ischaemic steal syndrome; venous hypertension) developed in four. The cumulative patency of PTFE AV fistulae was 74% after 1 year; 59% after 2 and 3 years and 47% after a follow-up of 5 years. Despite the high rate of complications the PTFE AV fistula has proved an acceptable technique in secondary access surgery for haemodialysis.

Is hysterectomy a risk factor for vaginal cancer?

Several recent case series have called attention to a possible association between previous hysterectomy and the subsequent development of vaginal cancer. To study this relationship, we compared 49 patients with vaginal cancer with 49 controls matched for age, race, and prior cervical dysplasia or neoplasia. Patients and controls were alike in terms of exposure to estrogens. Twenty-four patients (49%) had had prior hysterectomies, of which 13 (27%) were for benign disease. Similarly, 24 controls had a history of a hysterectomy. The matched-pairs odds ratio relating prior hysterectomy to vaginal cancer was 1.00 based on these data, with a 95% confidence interval of 0.47 to 2.12. In the subsample of women without a history of cervical disease, a similar odds ratio appeared. Although the study sample size did not permit exclusion of a twofold increase in risk, the statistical power to detect an actual odds ratio of 2.5 is 76%. At this level of statistical power, our data suggest that hysterectomy has a low probability of being a risk factor for vaginal cancer when age and cervical disease are controlled for. In the absence of such a relationship, screening for vaginal cancer does not appear to be necessary for women who have had a hysterectomy for benign disease.