Alveolar soft part sarcoma metastatic to small bowel mucosa causing polyposis and intussuseption

A report of alveolar soft part sarcoma metastatic to the small bowel is presented. Hematogenous metastases to the small bowel from primary tumors outside the abdominal cavity are uncommon, and most remain asymptomatic and are not discovered until autopsy. However, small bowel metastases can lead to intestinal obstruction, intussuseption or even perforation. While metastases to the small bowel have been described for other tumor types, including melanoma and lung cancer, this is extremely uncommon for sarcoma, especially alveolar soft part sarcoma. We describe a 42-year-old male with a long history of alveolar soft part sarcoma, metastatic to the lung and brain, who developed an intussuseption from metastases to the small bowel.

Renal atrophy secondary to chemoradiation treatment of abdominal malignancies

e15532

Background: Information on renal atrophy following abdominal chemoradiation (CRT) is limited. Methods: Patients who received concurrent CRT to the abdomen between 2002 and 2008 were identified for this study to evaluate change in renal size (RS) and function following CRT. Imaging and biochemical data were obtained prior to radiation (RT) and after RT in 6 month intervals. RS was defined by craniocaudal measurement on CT images. Renal function was assessed by serum creatinine and creatinine clearance (CrCl) using the Cockcroft-Gault formula. The primarily irradiated kidney (PK) was defined as the kidney that received the greater mean kidney dose. Results: Median patient age was 64 years (range 31–87), 51.5% of 130 patients were male. Primary disease sites were pancreas (61.5%), periampullary (16.1%), stomach (10.8%), gastroesophageal junction (10%), and retroperitoneum (1.5%). Median follow up was 9.4 months (range 0–55.4 mos). Median radiation dose was 50.4Gy (range 12.6- 55.8Gy). Mean dose to the PK was 18.6 Gy. Compensatory hypertrophy of the non-PK was not seen. Age, gender, hypertension, diabetes, smoking, pre-RT CrCl, and pre-RT RS were not associated with renal atrophy at 12 months post RT. Percent volumes of the PK receiving ≥10Gy (V10), 15Gy (V15), and 20Gy (V20) were significantly associated with renal atrophy at 12 months post RT (p=0.0030, 0.0041, and 0.0046 respectively). Conclusions: Significant detriments in PK size and renal function were seen following abdominal CRT. V10, V15, and V20 were associated with renal atrophy at 12 months post RT. These observations can assist with renal dose constraints in CRT treatment planning for patients at increased risk for renal complications.

[Table: see text]

No significant financial relationships to disclose.

Primary adenocarcinoma of the anus: a 22-year SEER population database analysis

e15072

Background: Most anal canal cancers consist of squamous cell carcinoma (SCCA). Adenocarcinoma (AdenoCa) is rare and accounts for approximately 10% of anal cancers. A search of the SEER database (1973 to 2005) was carried out to evaluate the pattern of radiation (RT) and surgical treatment. Impact of the treatment on overall survival (OS) was evaluated. Methods: The search of the SEER database revealed 1,008 pts who had pathologically confirmed anal cancers with either SCCA or AdenoCa. All pts had single diagnosis of anal cancer with localized disease without nodal involvement. Excluded were pts with unknown use of surgery or radiation. Kaplan-Meier estimates and Cox-Mantel model were used to calculate and compare survival rates. Results: Median age of all pts was 57 years (range 29 to 99). For the SCCA group, median age was 57 years (range 29 to 99), which was younger than the AdenoCa group with median age of 74 years (range 37 to 92). Among the 1,008 pts, 954 had SCCA (94.6%), and 54 had AdenoCa (5.4%). Within the SCCA group, 14 (1.5%) had abdominoperineal resection (APR) in combination with external beam RT, and 795 (83.3%) had RT only with non-APR local surgical treatment inclusive of excision. Remaining 145 SCCA pts (15.2%) had non-APR local surgical treatment only without RT or had no treatment. Within the AdenoCa group, 10 (18.5%) had APR in combination with external beam RT, and 21 (38.9%) had RT only with non-APR local surgical treatment inclusive of excision. Remaining 23 AdenoCa pts (42.6%) had non-APR local surgical treatment only without RT. None of the adenoCa pt had no treatment. OS at 10 years was superior for pts with SCCA vs. AdenoCa (64.9% vs. 25.8%, p<0.05). Among the SCCA subset, there was no signficant difference in the 10-yr OS between the APR versus the RT pts (71% vs. 65%, p=0.78). On the other hand, among the AdenoCa subset, pts who had APR had better 10-yr OS than RT pts (53.8% vs. 0%, p=0.03) Conclusions: For localized anal SCCA, RT yielded equivalent overall survival as compared to APR. On the other hand, pts with localized anal adenoCa appeared to do worse when APR was omitted. RT only without APR might not be sufficient treatment in these patients. This database lacks information on use of chemotherapy and local disease control. Omission of APR in pts with anal canal adenoCa should be cautiously weighed.

No significant financial relationships to disclose.

Epithelial-mesenchymal transition (EMT) and activated extracellular signal-regulated kinase (p-Erk) in surgically resected pancreatic cancer

BACKGROUND

EMT or transformation to the mesenchymal phenotype plays an important role in tumor invasion and metastasis. In vitro data suggest that mesenchymal transformation may correlate with the activation of PI3 kinase and Ras/Erk pathways. We investigated the expression of EMT markers (low E-cadherin, high fibronectin, and vimentin) and their association with p-Erk in resected pancreatic cancer.

METHODS

Clinical data/surgical specimens from 34 consecutive pancreatic cancer patients (pts) who underwent pancreatectomy were included. Immunohistochemical staining was performed on formalin-fixed paraffin-embedded tissues using monoclonal antibodies against vimentin, fibronectin, E-cadherin, and p-Erk. The results were correlated with clinicopathological parameters and survival. Survival analysis (log-rank test, Cox proportional hazard model), categorical data analysis (Pearson's chi-square, Fisher's exact test) and Kendall's tau were performed at a significance level of 0.05.

RESULTS

The patient population was formed from 13 males and 21 females, with a median age of 66 years (range 38-84 years); American Joint Committee on Cancer (AJCC) stage 1 (n = 2), 2 (n = 27), 3 (n = 5); histological grade 1 (n = 4), 2 (n = 13), 3 (n = 16), 4 (n = 1). Median survival was 15 months (95% CI: 11-24 months). Fibronectin overexpression correlated with the presence of vimentin (p = 0.0048) and activated Erk (p = 0.0264). There was a borderline association of fibronectin with worsening grade (p = 0.06). A negative association between vimentin and E-cadherin was noted (p = 0.0024). Increased fibronectin or vimentin and decreased E-cadherin correlated with poor survival.

CONCLUSION

EMT is associated with poor survival in surgically resected pancreatic adenocarcinoma. A correlation between activated Erk and fibronectin was identified that may open avenues for targeted therapy for this subgroup.

Geographic variation in soft tissue sarcoma patient follow-up

20518

Background: Of all newly diagnosed malignancies, 1% are soft tissue sarcomas (STS) and 59% of STS occur in the extremities. Most patients (80%) suffer recurrence within two years of potentially curative resection. Late recurrences at five years and beyond are not uncommon. Surveillance programs are designed to identify recurrence or new primaries early enough to positively impact survival and quality of life. Though published guidelines exist for extremity STS follow-up, adherence varies among physicians. Geographic factors were hypothesized to be potential predictors of this variation. Methods: The SSO membership (N=1592) was surveyed regarding postoperative STS surveillance using standardized clinical vignettes. Practice patterns were analyzed by metropolitan statistical area (MSA), managed care organization (MCO) penetration rate, and U.S. census region using repeated measures analysis of variance. The study endpoint was surveillance intensity. Results: Forty-five percent of SSO members (714) completed the survey; 343 (48%) performed sarcoma surgery. Of those who perform surgery, 318 (93%) follow their patients long-term. Mean follow-up intensity for the 12 modalities was highly correlated by tumor size, grade, and years after surgery. Controlling for tumor stage, grade, and year after surgery, geographic factors infrequently predicted surgeon self-reported surveillance intensity (p < .05). MSA was a significant predictor of office visit frequency. MCO penetration rate significantly predicted the frequency of urinalysis and site MRI. U.S. census region significantly predicted the frequency of LFTs. Two-way interaction effects were frequently significant. Few three-way interactions were examined due to sample size limitations. Conclusion: Geographic factors were generally not predictive of surgeon self-reported surveillance practice patterns for patients after curative-intent sarcoma surgery. Internship, residency, and fellowship locations may be important surgeon-specific aspects for future surveys.

No significant financial relationships to disclose.

Surgical resection for pancreatic cancer: Correlates of margin positivity

4097

Background: Positive surgical margins (PSMs) occur in 20–40% of pancreatic cancer cases treated surgically. PSMs are associated with local recurrence and poor survival. Objectives: Correlate clinicopathological parameters with PSMs in pancreatic cancer patients (pts) undergoing surgical resection. Methods: Clinicopathological data for 71 pts who underwent surgical resection for pancreatic cancer from 1999–2005 were obtained. Data including location, AJCC stage, tumor size, type of resection, use of neoadjuvant therapy, grade, morphology, perineural and angiolymphatic invasion were correlated with PSMs and survival using logistic regression, Fisher-exact test and log-rank test. Results: There were 29 males; median age was 65 years and median follow-up 14 months. Histologies were: adenocarcinoma (n = 67), sarcomatoid (n = 2), adenosquamous (n = 1) and mucinous cystadenocarcinoma (n = 1). Sites of disease were: head only (n = 56), head/uncinate (n = 7) and body/tail (n = 8). Surgical treatments included Whipple (n = 43), pylorus preserving Whipple (n = 18) distal pancreatectomy (n = 8) and total pancreatectomy (n = 2). Pathological AJCC stages were: ΙA (n = 4), ΙB (n = 2), ΙΙA (n = 14), ΙΙB (n = 48), and ΙΙΙ (n = 3). Neoadjuvant chemoradiation (CRT) was administered in 10 cases; 38 other pts received postoperative adjuvant therapy including CRT (n = 36)/chemotherapy only (n = 2). Overall median survival was 30 months and disease free survival 18.6 months. Nineteen cases (27%) had PSMs. PSMs were pancreatic transection line (n = 11), retroperitoneal margin (n = 9), bile duct (n = 1) and peripancreatic soft tissue (n = 2). On univariate analysis, PSM was associated with advanced AJCC stage (p = 0.02) and tumor location other than pancreatic head (p = 0.02). PSM rate was 10% (1/10) and 31% (18/61) for pts with/without neoadjuvant therapy, respectively (p = 0.27). On multivariate analysis, PSM correlated with AJCC stage (p = 0.04) and tumor location (p = 0.05). PSMs were not significantly related to survival or recurrence (p = 0.09). Conclusions: Advanced AJCC stage and tumor in pancreatic body/tail or uncinate are associated with PSMs following pancreatic cancer resection. Strategies to “downstage” with novel therapies prior to surgery should be investigated to facilitate a complete surgical resection.

[Table: see text]

Phase II study of gemcitabine, capecitabine and bevacizumab for advanced pancreatic cancer (APC) with ECOG PS 0–1

4117

Background: Gemcitabine and capecitabine chemotherapy may result in improved survival for patients (pts) with APC and good performance status. Preclinical synergy between these agents and bevacizumab (BV) and encouraging results of angiogenic inhibition in PC formed the rationale for this study. Endpoints: Primary: progression-free survival (PFS), Secondary: response rates, overall survival, toxicity assessment and quality of life (QOL). Methods: Pts with APC (inoperable stage III or IV), ECOG PS 0–1, normal organ and marrow function were eligible. Schema: Avastin 15 mg/kg, q 21 days; capecitabine 650 mg/m2 bid x 14 days, both starting day 1; gemcitabine 1000 mg/m2 days 1 & 8; cycles repeated q 21 days. The study has a power of 0.8 at 0.05 significance level to determine if PFS is ≥ 4.1 months with study regimen (2.5 months with gemcitabine). Results: Twenty-six (of planned 34) pts have been enrolled; 22 pts were evaluable for response (4 too early). Median age 64 years (range 38–79), 12 are male, one stage III, 25 stage IV. Median of 3 cycles/pt (range 1–16) with a total 105 cycles. Responses (RECIST): 7 partial responses (32%; 95% CI: 14%-55%), 12 stable disease, 3 progressive disease. The median PFS is 8.2 months (C.I. 4.7–11.3). Estimated median overall survival is 9.1 months (C.I. 8.3-undetermined). Five% improvement in QOL occurred in 7/15 pts (95% C.I: 21–73%). Grade 3 toxicities: proteinuria (n=1), neutropenia (n=2), thrombocytopenia (n=1), pulmonary embolism (n=2), nausea (n=2), emesis (n=2), diarrhea (n=1), infection (n=1), anemia (n=1) and hemorrhage (n=1). Grade 4 toxicities: neutropenia, cerebrovascular accident and diarrhea (n=1, each). One treatment related death occurred (hemorrhage). Conclusion: In this interim analysis, encouraging response rates, PFS and QOL improvement occurred with the addition of BV to gemcitabine and capecitabine. Toxicity profile is acceptable. Accrual is ongoing. Acknowledgement: Study is supported by a grant from NCCN and by Genentech.

[Table: see text]

Is glucagonoma of the pancreas a curable disease?

BACKGROUND

Glucagonomas are rare neuroendocrine tumors of the pancreas. Because of its rarity, its natural history is not well understood.

AIM

We evaluated the natural history of glucagonomas treated at a tertiary care cancer center.

METHODS

A retrospective analysis of 12 patients during 1970 to 2000 was performed. Six patients (50%) had a tumor located in the head of the pancreas.

RESULTS

Abdominal pain (83%) and weight loss (75%) were the most common symptoms. Median tumor size was 6 cm (range 0.04-10). Seven patients (58%) had liver metastases. Five patients (42%) underwent curative resection. Overall median survival was 66 mo, and 5-yr overall survival was 66%. Five-yr overall survival was 83% for patients who had resection versus 50% for the non-resected patients (p = 0.04). Patients who were disease-free had a complete resection of the primary tumor and no liver involvement.

CONCLUSIONS

Glucagonomas generally present with liver metastases at the time of diagnosis. Cure is only possible if the disease is localized and completely resected.

A phase I study of fever-range whole body hyperthermia (FR-WBH) in patients with advanced solid tumours: correlation with mouse models

Various studies in animal tumour models have revealed the potential of fever-range whole body hyperthermia (FR-WBH) to be used in cancer therapy. To determine the safety of FR-WBH treatment in the clinic, patients with advanced solid tumours were heated in the outpatient setting to 39-39.5 degrees C for 3 or 6h, or 39.5-40 degrees C for 6h using the Heckel-HT 2000 apparatus. These WBH treatments were well tolerated, with no significant adverse events related to cardiac, hepatic, renal or pulmonary systems. In the majority of patients, flow cytometric analysis of peripheral blood leukocyte populations indicated that there were transient decreases in the number of circulating T lymphocytes and a concomitant decrease in the number of L-selectin positive lymphocytes in the peripheral blood. These findings closely mimic the affects seen previously in pre-clinical murine studies in which this same fever-like treatment was shown to inhibit tumour growth. These studies have established the safety of this treatment and will allow for future clinical trials where application of FR-WBH treatment can be combined with other anti-cancer therapies, including immunotherapy and chemotherapy.

Image analysis for quantitation of solid tumor drug sensitivity

BACKGROUND

A method of assessing chemosensitivity of tissue has been described by Rotman et al. The aim of this study was to use image analysis to provide a more rapid and quantitative means of assessing drug effect on tissue proliferative capacity.

METHOD

Fluoropyrimidine sensitive Ward rat colon adenocarcinoma tumor was implanted onto collagen impregnated cellulose fibers suspended on metal grids at an air-fluid interface and kept in a 95% air/5% CO2 incubator at 37 degrees C. The fluorescent microscopic image captured by a silicon intensified target (low light detecting) camera and linked to an image processing unit was measured for fluorescent brightness and tumor image area. Blinded 5-Fluorouracil (5-FU) drug treatment was begun 8 days after tumor explantation on the collagen-cellulose matrix. Tumor image area and fluorescent brightness were measured at 24 h pretreatment, 48 h posttreatment, and at 48 h post drug removal.

RESULTS

Nontreated tumor cultures demonstrated an increase in area and fluorescent brightness with time following tumor implantation on the collagen gel. Dose responsiveness was seen with increasing concentrations of 5-FU. At the highest clinically achievable concentration of 5-FU (500 microM), there was a 39% decrease in area compared with the nontreated group, 113%. Linear dose responsiveness was not demonstrated between 50 and 150microM 5-FU.

CONCLUSIONS

Fluoropyrimidine activity was demonstrated with the implemented image analysis system. The in vitro tumor sensitivity to FU using collagen gel was consistent with responsiveness of tumors in vivo borne by rats.

Hepatic lymphatic mapping: a pilot study for porta hepatis lymph node identification

The status of the porta hepatis lymph nodes in patients with hepatic metastases from colorectal cancer affects their prognosis and management. Lymphatic mapping with isosulfan blue dye is well established in breast cancer and melanoma. An animal model consisting of three dogs receiving general anesthesia was utilized. Each dog underwent a laparotomy and increasing doses of isosulfan blue dye were injected into the right medial segment of the liver. Intraoperatively, the presence of blue dye in the porta hepatis region was determined and the lymph node identified. Continuous physiological monitoring was performed. Serum determination of liver function tests, amylase levels, and white blood cell count were performed preoperatively and on postoperative days 1, 2, 4, and 7. The animals were sacrificed on day 7. A portal lymph node was identified in each case and there was no perioperative morbidity or mortality. There were no significant alterations in blood pressure or heart rate in the animals. There was a dose-responsive decrease in the O2 saturation as measured by transcutaneous monitoring, but arterial blood gas analysis showed that pO2 levels remained stable. There were no significant changes in the liver function tests, amylase levels, or white blood cell counts. There was a small increase in alkaline phosphatase, which normalized by postoperative day 7. Hepatic injection of isosulfan blue dye appears to be safe and effective in identifying porta hepatis lymph nodes in the animal model and sets the basis for further study in human subjects.

Clear cell sarcoma: the Roswell Park experience

BACKGROUND AND OBJECTIVES

Clear cell sarcoma of the tendons and aponeuroses (CCSTA) is an aggressive, rare soft-tissue tumor with approximately 300 reported cases. Although it appears to be histogenetically related to melanoma, its clinical behavior resembles soft tissue sarcoma with a propensity for lymph node metastases. We report our experience at a tertiary cancer center.

METHODS

Eight cases of CCSTA evaluated at Roswell Park Cancer Institute between 1970 and 1998 were reviewed retrospectively. Patient data analyzed included patient age, gender, anatomic location, size of tumor, development of local, regional and distant recurrence, and patient status at last follow up.

RESULTS

Six of eight patients were alive at 2 years, while three of seven patients were alive at 5 years. Of the patients alive with no evidence of recurrence, two had tumors of less than 2 cm, and the remaining patient had incomplete information regarding tumor size. Five patients recurred within 2 years of definitive surgical management. Four had tumors > 5 cm. All five patients progressed to metastatic disease at a median follow up of 20 months (range 1-108 months) following definitive surgical management and all eventually died of their disease at a median of 3 months (range 0-24 months) from presentation with metastatic disease. Four of five patients with lesions > 5 cm received adjuvant chemotherapy with intent to cure, but all eventually died of disease at 4, 22, 34, and 41 months from initial presentation.

CONCLUSIONS

CCSTA is an aggressive tumor of the soft tissues. Early recognition and management are associated with an excellent long-term prognosis. Tumors greater than 5 cm warrant aggressive surgical management and treatment, and are at high risk of the development of distant disease. Aggressive multiagent chemotherapy appeared to have no impact on outcome. Other adjuvant therapeutic options including immunotherapy should be investigated.

Malignant fibrous histiocytoma: an institutional review

BACKGROUND

A thorough understanding of malignant fibrous histiocytoma (MFH), the most common subtype of soft tissue sarcoma, will lead to improved histologic-specific protocols.

METHODS

126 patients with histologically confirmed MFH were analyzed. The median follow-up was 42 months (range 1-233 months).

RESULTS

Overall survival was 58% at 5 years and 38% at 10 years. Grade significantly influenced prognosis, with 10-year survival of 90%, 60%, and 20% for low, intermediate, and high grade tumors, respectively (p = 0.0007). Distant metastases at initial presentation (p = 0.0002) and size of the primary tumor (p = 0.0007) influenced outcome. Neither anatomic site nor depth of the primary tumor were significant prognostic factors. Positive microscopic margins were associated with a decreased disease-free survival (p = 0.006).

CONCLUSIONS

Tumor grade, size, and distant metastases at initial presentation remain the most important prognostic factors for MFH. Resection with negative microscopic margins decreased the incidence of local recurrence.

Risk factors for nodal recurrence after lymphadenectomy for melanoma

BACKGROUND

The risk and outcome of regional failure after elective and therapeutic lymph node dissection (ELND/TLND) for microscopically and macroscopically involved lymph nodes without adjuvant radiotherapy were evaluated.

METHODS

Retrospective melanoma database review of 338 patients (ELND 85, TLND 253) from 1970 to 1996 with pathologically involved lymph nodes.

RESULTS

Regional recurrence occurred in 14% of patients treated with ELND (n = 12) and 28% of patients treated with TLND (n = 72; P = .009). Risk factors associated with nodal recurrence were advanced age, primary lesion in the head and neck region, depth of the primary lesion, number of involved lymph nodes, and extracapsular extension (ECE). For each nodal basin, the ELND group had a lower incidence of recurrence than the TLND group. The TLND group had larger lymph nodes, greater number of involved lymph nodes, and a higher incidence of ECE. The 10-year disease-specific survival was 51% vs. 30% for ELND and TLND, respectively (P = .0005). Nodal basin failure was predictive of distant metastasis, with 87% developing distant disease compared with 54% of patients without nodal recurrence (P < .0001). Of six patients who underwent a second dissection after isolated nodal recurrence, five patients have had a median disease-free interval of 79 months.

CONCLUSIONS

After ELND or TLND, patients who have a large tumor burden (thick primary melanoma, multiply involved lymph nodes, ECE), advanced age, and a primary lesion located in the head and neck have a significantly increased likelihood of relapse and a decreased survival. Few patients present with an isolated nodal recurrence, but the majority can be salvaged by a second dissection.

Evolution of sentinel lymph node biopsy for melanoma at a National Cancer Institute-designated cancer center

BACKGROUND

Sentinel lymph node biopsy (SLNB) has rapidly evolved into the standard of care for clinically node-negative melanoma. Since adopting sentinel lymph node (SLN) technology in 1993, we have periodically reviewed our institution's results and made several modifications.

METHODS

From January 1993 to December 1998, 182 patients with clinically node-negative primary cutaneous melanoma underwent SLNB. Charts were retrospectively reviewed and assessed for the technique for the identification of the SLN, the pathologic analysis, and the use of intraoperative frozen section.

RESULTS

The accuracy of SLN identification improved from 91% to 100% with the combination of isosulfan blue dye and radiolabeled colloid over isosulfan blue dye alone. Routine versus selective lymphoscintigraphy identified 7 in-transit SLNs and increased detection of dual nodal basin drainage (15%-27%). Identification of micrometastases in the SLN increased from 14% to 24% after a modification of pathologic evaluation. The positive SLN was the only involved node in most patients (80%). Intraoperative frozen section had a sensitivity of 58% and was of benefit in only 13 of 124 patients (10%).

CONCLUSIONS

Several modifications to the identification of the SLNs and the detection of metastatic melanoma have improved our outcome with SLNB. A careful, periodic review of results to identify areas for improvement at each institution is crucial to the success of SLNB for melanoma.

Gastrointestinal stromal tumors: current diagnosis, biologic behavior, and management

Gastrointestinal stromal tumors (GIST) are rare tumors of the gastrointestinal (GI) tract that arise from primitive mesenchymal cells. GISTs occur throughout the GI tract but are usually located in the stomach and small intestine. The majority of GISTs are immunohistochemically positive for c-kit protein (CD 117) and CD34. GISTs express a heterogeneous clinical course not easily predicted by standard pathological means. The most important prognostic factors are size > 5 cm, tumor necrosis, infiltration and metastasis to other sites, mitotic count > 1-5 per 10 high-powered fields, and most recently, mutation in the c-kit gene. Surgical resection remains the mainstay of treatment, as chemotherapy and radiation are ineffective. Long-term follow-up is imperative, as recurrence rates are high.

Pathologic and clinical features influencing outcome of thin cutaneous melanoma: correlation with newly proposed staging system

The incidence of malignant melanoma is increasing. Because of increased awareness, early recognition of malignant melanoma has become more common. In 1997, a new staging system for cutaneous melanoma was proposed, with reclassification of thin melanoma < 1 mm, with and without ulceration. This report evaluates the pathologic and clinical features of thin melanomas influencing recurrence and survival from a tertiary cancer center in an attempt to correlate findings with the proposed staging system. A review of the Roswell Park Cancer Institute tumor registry identified 352 patients with thin cutaneous melanomas (< 1.0 mm) seen during an 18-year period ending August 30, 1998. Overall survival was 93 and 87 per cent at 5 and 10 years, respectively. Disease-free survival was 94 and 93 per cent at 5 and 10 years, respectively. Local recurrence occurred in 3 per cent of patients, regional recurrence in 3 per cent, and metastatic disease in 3 per cent, for an overall recurrence of 7 per cent, with a median follow-up of 118 months. Only the presence of ulceration was a significant prognostic factor for recurrence by both univariate and multivariate analysis. Failure rates (any recurrence) by Clark levels I, II, and III/IV were 3, 5, and 10 per cent, respectively (P = 0.14). Failure rates by tumor thickness (mm), for 0.0-0.24, 0.25-0.49, 0.50-0.74, and 0.75-0.99 were 3, 4, 7, and 10 per cent, respectively (P = 0.49). Ten-year disease-free survival for ulceration versus no ulceration was 40 and 94 per cent, respectively (P < 0.0001). We conclude that thin cutaneous melanoma carries an excellent prognosis with appropriate treatment. Our findings support inclusion of ulceration in a new staging system. Lesions 0.76 to 0.99 mm and Clark level III and IV may warrant close observation as a separate subgroup.

Current follow-up strategies after potentially curative resection of extremity sarcomas: results of a survey of the members of the society of surgical oncology

BACKGROUND

The follow-up of patients after potentially curative resection of extremity sarcomas has significant clinical and fiscal implications. However, the ideal postoperative surveillance regimen for these uncommon neoplasms remains ill-defined. This study was designed to determine the current follow-up practices of a large, diverse group of physicians who care for sarcoma patients.

METHODS

The 1592 members of the Society of Surgical Oncology (SSO) were surveyed regarding their follow-up practices with a detailed questionnaire mailed in 1997. Information regarding frequency of follow-up testing was requested for extremity sarcoma patients treated for cure based on 4 vignettes: low grade lesion </= 5 cm and > 5 cm and high grade lesion </= 5 cm and > 5 cm. Respondents were asked to indicate the number of office visits, laboratory tests and imaging studies performed annually during the first 5 years and the 10th year after surgery.

RESULTS

Forty-five percent (716 of 1592) completed the survey. Of the 343 respondents who performed sarcoma surgery, 318 (93%) also provided long term postoperative follow-up for their patients. Ninety-four percent of respondents (295 of 318) were trained in general surgery and 5% (15 of 318) completed orthopedic residencies. Ninety-one percent (291 of 318) were also fellowship trained (80% in surgical oncology). Sixty-three percent (201 of 318) were in academic practice. Routine office visits and chest X-ray (CXR) were the most frequently performed items for each of the years. The frequency of office visits and CXR increased with tumor size and grade and decreased with postoperative year. Complete blood count and liver function tests were the most commonly ordered blood tests, but many respondents did not order any blood tests routinely. Imaging studies of the extremities were performed on the majority of patients with large (> 5 cm) low grade lesions and on both large and small high grade lesions during the first postoperative year.

CONCLUSIONS

Postoperative sarcoma surveillance strategies utilized by members of the SSO rely most heavily on office visits and CXR. Tumor grade, tumor size, and postoperative year affect surveillance intensity.

Clinical importance of late recurrence in soft-tissue sarcomas

BACKGROUND AND OBJECTIVES

Soft-tissue sarcomas (STS) represent a diverse histologic group of malignancies at risk for local and distant failure. We studied the impact of late (5 or more years) vs. early recurrence (less than 5 years) on subsequent outcome.

METHODS

Four hundred sixty-eight patients with STS treated between 1962 and 1992 were evaluated for late (n = 39; 8%) or early (n = 253; 54%) recurrence. Clinical and pathologic factors were reviewed. Survival data were analyzed by the Kaplan-Meier method and the log-rank test.

RESULTS

Of the 39 patients with a late recurrence (median follow-up 156 months), 18 patients had local recurrence, 7 patients developed distant recurrence, and 14 patients had local and distant recurrence. Thirty patients with late local and/or distant recurrence underwent complete or wide excision (n = 16), amputation (n = 4), or local resection (n = 10). The overall 5-year survival rate following late recurrence was 61%. The 5-year overall survival rate was statistically better for patients with a late local recurrence alone than for patients with distant failure, 94% vs. 36%, respectively (P = 0.003). Neither the site of the primary STS, age, primary margin status, nor histology had any effect on subsequent local or distant failure and subsequent survival.

CONCLUSIONS

These data suggest that an aggressive approach is appropriate in patients who present with late recurrence (more than 5 years) following treatment of the primary STS. Impressive survival rates can be achieved in the treatment of local recurrences.

Nodal basin recurrence following lymph node dissection for melanoma: implications for adjuvant radiotherapy

PURPOSE

To analyze patterns of failure in malignant melanoma patients with lymph node involvement who underwent complete lymph node dissection (LND) of the nodal basin. To determine prognostic factors predictive of local recurrence in the lymph node basin in order to select patients who may benefit from adjuvant radiotherapy.

METHODS AND MATERIALS

A retrospective analysis of 338 patients undergoing complete LND for melanoma between 1970 and 1996 who had pathologically involved lymph nodes was performed. Mean follow-up from the time of LND was 54 months (range: 12-306 months). Lymph node basins dissected included the neck (56 patients), axilla (160 patients), and groin (122 patients). Two hundred fifty-three patients (75%) underwent therapeutic LND for clinically involved nodes, while 85 patients (25%) had elective dissections. Forty-four percent of patients received adjuvant systemic therapy. No patients received adjuvant radiotherapy to the lymph node basin.

RESULTS

Overall and disease-specific survival for all patients at 10 years was 30% and 36%, respectively. Overall nodal basin recurrence was 30% at 10 years. Mean time to nodal basin recurrence was 12 months (range: 2-78 months). Site of nodal involvement was prognostic with 43%, 28%, and 23% nodal basin recurrence at 10 years with cervical, axillary, and inguinal involvement, respectively (p = 0.008). Extracapsular extension (ECE) led to a 10-year nodal basin failure rate of 63% vs. 23% without ECE (p < 0.0001). Patients undergoing a therapeutic dissection for clinically involved nodes had a 36% failure rate in the nodal basin at 10 years, compared to 16% for patients found to have involved nodes after elective dissection (p = 0.002). Lymph nodes larger than 6 cm led to a failure rate of 80% compared to 42% for nodes 3-6 cm and 24% for nodes less than 3 cm (p < 0.001). The number of lymph nodes involved also predicted for nodal basin failure with 25%, 46%, and 63% failure rates at 10 years for 1-3, 4-10, and > 10 nodes involved (p = 0.0001). There was no significant difference in nodal basin control in patients with synchronous or metachronous lymph node metastases, nor in patients receiving or not receiving adjuvant systemic therapy. Nodal basin failure was predictive of distant metastasis with 87% of patients with nodal basin recurrence developing distant disease compared to 54% of patients without nodal failure (p < 0.0001). On multivariate analysis, number of positive nodes and type of dissection (elective vs. therapeutic) were significant predictors of overall and disease-specific survival. Size of the largest lymph node was also predictive of disease-specific survival. Site of nodal involvement and ECE were significant predictors of nodal basin failure.

CONCLUSIONS

Malignant melanoma patients with nodal involvement have a significant risk of nodal basin failure after LND if they have cervical involvement, ECE, >3 positive lymph nodes, clinically involved nodes, or any node larger than 3 cm. Patients with these risk factors should be considered for adjuvant radiotherapy to the lymph node basin to reduce the incidence of nodal basin recurrence. Patients with nodal basin failure are at higher risk of developing distant metastases.

Primary extremity sarcoma: what is the appropriate follow-up?

BACKGROUND

Our objective was to evaluate the effectiveness of follow-up tests for detecting first local and distant recurrences in patients with primary extremity soft tissue sarcoma.

METHODS

We retrospectively analyzed all adult cases of primary extremity soft tissue sarcoma (n = 174) treated between 1982 and 1992. Patients were observed every 3 months for 2 years, every 4 months the third year, every 6 months the next 2 years, and annually, thereafter. Each visit consisted of taking the patient's history, a physical examination, a complete blood count, a blood chemistry panel, and a chest x-ray. For high-grade tumors, the primary site was imaged annually when clinically appropriate.

RESULTS

Of 141 patients who were assessable, 29 patients developed local recurrence and 57 developed distant recurrence. All but one of the local recurrences was detected on the basis of an abnormal physical examination. Of the 29 patients who developed local recurrence, 25 were resected. Distant metastases were detected because of symptoms in 21 cases. Of the 36 asymptomatic lung recurrences, 30 were detected by follow-up chest x-ray. Of the 36 asymptomatic lung recurrences, 24 patients underwent metastasectomy. The positive and negative predictive values of surveillance chest x-ray were 92% and 97%, respectively. Laboratory testing never led to the detection of recurrence.

CONCLUSIONS

Close surveillance by clinical assessment and chest x-ray is appropriate for follow-up observation of patients with primary extremity soft tissue sarcoma.

Large, deep high-grade extremity sarcomas: when is a myocutaneous flap reconstruction necessary?

The surgical treatment of large, deep high-grade extremity soft tissue sarcomas frequently produces a significant tissue defect. In addition, the management of the surgical wound is often further complicated by preoperative radiation or adjuvant therapies. The use of either pedicled or free myocutaneous flaps allows for more rapid and predictable wound healing in this situation. Myocutaneous flaps provide well-vascularized coverage of lost tissue volume, exposed vital structures, and prosthetic reconstruction materials. When harvested from unirradiated sites, flap coverage can overcome the detrimental effects of radiation therapy and chemotherapy on postoperative wound healing. Reconstruction of the soft tissue defect may also improve patient satisfaction with aesthetic issues. The use of innervated myocutaneous flaps can even address the functionality of the extremity following resection of major muscle groups. Myocutaneous flaps are an extremely versatile option for reconstruction in the treatment of large, deep high-grade extremity soft tissue sarcomas.

Paraesophageal omental hernia mimics pleural lipomatous tumor

Paraesophageal omental herniation (POH) is uncommon. CT scan and MRI are complementary in diagnosis. We present a posterior mediastinal mass in a 43 year old male with a history of myxoid liposarcoma raising the suspicion of latent secondary tumor. Subsequently, at thoracotomy he was found to have a POH. Differential diagnosis, work-up and surgical approach are discussed.

Accuracy of pathologic techniques for the diagnosis of metastatic melanoma in sentinel lymph nodes

BACKGROUND

Sentinel lymph node (SLN) biopsy can accurately predict the presence of metastatic melanoma (MM) and has been used to identify patients with occult metastases. We present an analysis of the sensitivity and specificity of standard pathological techniques including intraoperative frozen section, permanent section, and immunohistochemistry in diagnosing MM within the SLN.

METHODS

Sixty-nine consecutive patients with primary malignant melanoma thickness of >1.0 mm or thinner lesions invading the reticular dermis (Clark level IV) who underwent SLN biopsy were reviewed. Lymph nodes were examined intraoperatively by frozen section (FS), permanent section (H&E), and by immunohistochemistry (IH) for S-100 protein and HMB45.

RESULTS

MM was found in 14 of 69 cases (20%). Permanent section H&E was performed in all cases, FS in 64 cases, and IH in 65 cases. FS analysis diagnosed MM in 4 of 14 cases (29%), was suspicious in 2 of 14 (14%), and falsely negative (FN) in 8 of 14 (57%) ultimately found to be positive with further workup. Within the FN group, MM was identified on review of the original FS slides in 3 of 8 cases (38%). Furthermore, within the FN group, the remaining 5 cases were identified as positive for MM by either permanent and/or deeper H&E sections and IH. IH alone with permanent H&E sections would have diagnosed MM in only 8 of 10 cases (80%) that were FS negative or suspicious. In no cases was MM identified by IH alone with the permanent and deeper H&E sections being negative. It is noteworthy that no false-positive cases were identified.

CONCLUSIONS

Intraoperative FS has low sensitivity in identifying MM within the SLN. IH alone does not increase the diagnostic yield. A combination of permanent H&E sections with deeper levels and S-100 and HMB45 IH dramatically increases the overall diagnostic sensitivity of SLN biopsy. Definitive diagnosis should await permanent H&E sections and IH staining.

Effective follow-up strategies in soft tissue sarcoma

The value of surveillance for detection of recurrences in patients with soft tissue sarcoma (STS) after definitive surgical resection of the primary tumor is based on the premise that early recognition and treatment of local or distant recurrence can prolong survival. Surveillance strategies should meet the criteria of easy implementation, accuracy, and cost-effectiveness. Although guidelines have been proposed for follow-up of patients with STS, there are few data in the medical literature on the effectiveness of these recommendations. We reviewed the effectiveness of a surveillance program for primary extremity STS in an effort to provide an evidence-based rationale for follow-up of STS. We concluded that clinical assessment of patient symptoms, chest X-ray imaging, and physical examination are effective strategies for follow-up of extremity STS. Chest X-ray imaging also appears to be cost-effective, at least for high-grade extremity STS. Imaging of the primary extremity site by computed tomography (CT) scan or magnetic resonance imaging (MRI) on an annual basis and routine laboratory blood tests were ineffective strategies for recurrence detection. However, certain patient characteristics such as body habitus, previous radiation therapy, and location of the primary tumor site may require the use of CT scans and MRI for adequate clinical assessment. The role of specific surveillance strategies for recurrence detection for sarcomas of the trunk, head and neck, retroperitoneum, and viscera has yet to be defined.

Intraoperative determinants of unresectability for patients with colorectal hepatic metastases

BACKGROUND

Intrahepatic and extrahepatic factors are utilized by the surgeon in the decision-making process for the performance of hepatic resection for patients with colorectal metastases. Accurate preoperative and intraoperative staging are mandatory to avoid unnecessary surgery. In this report the intraoperative determinants of hepatic unresectability were evaluated.

METHODS

This was a retrospective review of medical records from January 1985 to March 1996 of 62 patients with colorectal hepatic metastases who at the time of exploratory laparotomy were deemed to have unresectable disease based on intrahepatic or extrahepatic factors. The stage of the primary tumor, disease free interval, preoperative carcinoembryonic antigen, computed tomography portography, intraoperative ultrasound, and assessment of intrahepatic and extrahepatic tumor extension were evaluated.

RESULTS

Intraoperative determination of the extent of required hepatic resection, including trisegmentectomy (9 patients; 15%) and total hepatectomy (10 patients; 16%), accounted for the majority of unresectable patients. Patients with > 4 metastases (8 patients; 13%) and satellitosis (6 patients; 10%) accounted for 23% of unresectable patients. Four patients had extensive nonmalignant hepatic parenchymal disease precluding resection. Thorough abdominal exploration revealed extrahepatic disease in 13 of 62 patients (21%). Routine periportal/celiac lymph node biopsies revealed metastases in an additional 12 patients (19%), 7 of whom (11%) had only periportal/celiac lymph node metastases.

CONCLUSIONS

A meticulous abdominal exploration prior to hepatic resection for patients with colorectal metastases is essential to identify those patients with extrahepatic disease. Periportal and celiac lymph nodes commonly are involved by tumor. Therefore, routine periportal/celiac lymph node biopsies should be performed in the absence of other extrahepatic disease.

Novel cellular determinants for reversal of multidrug resistance in cells expressing P170-glycoprotein

The newly synthesized calcium channel blocker, Ro44-5912, significantly potentiates doxorubicin (Dox)-induced cytotoxicity at non-cytotoxic concentrations in Dox-resistant human ovarian cell line, A2780/DX5, overexpressing P170-glycoprotein (Pgp). Induction of DNA single- and double-strand breaks (ssbs and dsbs) was measured using alkaline elution and constant-field gel electrophoresis (CFGE) assays. The results indicate that potentiation of the cytotoxicity of Dox by Ro44-5912 was accompanied by significant increases in both, Dox-induced DNA ssbs and dsbs in the resistant cells. Pulsed-field gel electrophoresis (PFGE) analysis showed that Dox induced DNA fragments in the 50-800 kilobase (kb) and 0.8-5.7 megabase (Mb) ranges. The majority of the newly synthesized DNA fragments were in the 50-800 kb range. Ro44-5912 treatment resulted in significant potentiation of DNA fragmentation in the 50-800 kb range with a minor increase in 0.8-5.7 Mb DNA fragments, suggesting that the modulator functions by potentiating nascent DNA fragmentation in the resistant cells. Exposure to Dox with Ro44-5912 was associated with a prolonged blockage of cells in the S-phase. In contrast, exposure to Dox alone resulted in temporary blockage of cells in G2/M phase (approximately 24 h) followed by restoration of cell proliferation and normal DNA histograms at 48 h after 2 h drug exposure. Incorporation of BrdUrd by flow cytometric analysis was inhibited by Dox in the presence of Ro44-5912, showing that there is a block of DNA replication. An increased damage in newly synthesized DNA could concur with a blocked DNA replication. Moreover, slowing progression through the S-phase in cells exposed to Dox in combination with Ro44-5912 is accompanied by increased sensitivity of Dox poisons, indicating a correlation of specific S-phase perturbation with the reversal of Dox resistance by Ro44-5912 in cells expressing Pgp. The results suggest that drug-induced augmentation of nascent DNA fragmentation and specific cell-cycle perturbation are potentially important molecular determinants for reversal of multidrug resistance in addition to restoration of intracellular drug retention.

Sentinel node biopsy in breast cancer

BACKGROUND

Sentinel lymph node biopsy (SNB) in breast cancer may be used in place of axillary lymph node dissection (ALND) if SNB accurately stages the axilla. This study assessed the success and accuracy of axillary SNB with isosulfan blue (ISB) and technetium-99 sulfur colloid (TSC) compared to ALND.

METHODS

Forty-two women with T1 or T2 breast cancer underwent SNB and ALND. Sixty to 90 minutes before anesthetic induction, a mixture of 3 mL ISB and 1 mCi TSC was injected around the primary cancer or prior biopsy site. Intraoperatively, the SLN was identified using a gamma detector (Neoprobe 1000) or by visualization of the blue-stained lymph node and afferent lymphatics. The SLN was excised separately, and a level I/II ALND was completed. The histologic findings of the axillary contents and SLN were compared.

RESULTS

An axillary SLN was found in 38 of 42 (90%) cases. SLN localization rate and predictive value were the same for women who had and those who had not undergone excisional biopsy before the date of SNB. Fifteen of 42 (36%) patients had lymph node metastases. The SLN was positive in all women with axillary metastases (negative predictive value, 100%).

CONCLUSIONS

If confirmed by larger series, a negative SNB may eliminate the need for ALND for select women with breast cancer.

Complications and outcome of external hemipelvectomy in the management of pelvic tumors

BACKGROUND

Although the technique of external hemipelvectomy has been adequately described, little is known about its complications and late results.

DESIGN

Retrospective review of 68 external hemipelvectomies performed at our Institute between 1973 and 1994.

MATERIALS AND METHODS

Eleven patients had bone tumor; 39 patients, soft-tissue sarcoma; seven patients, melanoma; 10 patients, squamous cell carcinoma; and one patient, giant neurofibroma. In 48 (71%) patients, the intent was curative. In 17 cases, the hemipelvectomy was extended.

RESULTS

Postoperative complications occurred in 36 (53%) patients, including flap necrosis in 11 (16%), wound infection in 24 (35%), and other complications in 12 (18%). Four (6%) patients died postoperatively. The average hospital stay after curative versus palliative resection was 39 versus 24 days. Only three (5%) patients were able to use a prosthesis, whereas 55 (81%) used crutches, six (9%) remained wheelchair bound, and four patients (6%) spent most of the time in bed. Local recurrence occurred in 35% of the patients. The estimated 5-year survival for curatively resected patients was 21%.

CONCLUSIONS

External hemipelvectomy is a procedure with considerable morbidity and is indicated for only a minority of far-advanced tumors. It offers a chance of palliation and possibly cure when lesser surgical options have been exhausted.

Cellular heterogeneity in DNA damage and growth inhibition induced by ICI D1694, thymidylate synthase inhibitor, using single cell assays

Heterogeneity in the response of the HCT-8 (human ileocecal adenocarcinoma) tumor cell line to a new thymidylate synthase inhibitor, ICI D1694, was investigated in terms of induction of DNA single-strand breaks and cytotoxicity, applying the single cell alkaline gel (SCG) electrophoresis assay and the individual colony formation assay (iCFA), respectively. ICI D1694 induced maximal total DNA single-strand breaks 24 hr after a 2-hr drug exposure with incomplete repair by 72 hr. The level of DNA damage was concentration dependent and paralleled cellular growth inhibition in vitro. The proportion of cells with DNA damage and the extent of DNA single-strand breaks increased with drug concentration. At 1 microM ICI D1694 (IC95), a significant level of DNA damage was detected in 58% of the cells; however, 25% of the cells had little or no damage. Using the iCFA system, it was observed that with 1 microM ICI D1694, only 2.6% of the seeded cells maintained a colony growth rate similar to that of the control colonies, and 22% of the cells were growing significantly more slowly. In conclusion, the SCG assay and the iCFA identified subpopulations of cells that were unaffected by ICI D1694. Although these cells represented only a small proportion of the total cell population, this phenomenon of heterogeneity in response to ICI D1694 might limit its therapeutic efficacy.

The use of marginal donors for liver transplantation. A retrospective study of 365 liver donors

A total of 365 donor hepatectomies performed between May 1985 and March 1990 were reviewed and analyzed retrospectively to identify risk factors associated with poor graft function and to study the outcome of grafts retrieved from "marginal" donors. The donor mean age was 27.1 years (8-69 years). Mean ICU donor stay was 2.7 days (range 0 to 18 days), and the mean ischemic time was 8.6 hr (range 3 to 22 hr). The pancreas was retrieved in 39 donors. Donor's weight above 100 kg was the only variable found to be associated with both significantly increased 3-month graft loss (P less than 0.01) and early hepatocellular damage--AST or ALT greater than 2000 U/ml, 1st day posttransplant (P less than 0.02). Prolonged stay in the ICU (greater than 3 days), although associated with a significantly increased rate of hepatocellular damage (P less than 0.05), did not affect early graft survival. A systolic blood pressure less than 90 mmHg despite the use of high-dose dopamine (greater than 15 micrograms/mg/min), but not each of these variables itself, was also associated with a significantly increase rate of hepatocellular damage (P less than 0.001). All other variables, including age greater than 50, ischemic time greater than 12 hr, combined liver-pancreas procurement, and liver function test abnormalities, did not affect the outcome. We conclude that extending our limits to accept donors of the higher age group and those who have moderately abnormal liver function tests or a prolonged ischemic time will not jeopardize our results. It is suggested to perform liver biopsy in overweight donors during the retrieval to prevent using grafts with severe fatty infiltration. It is hypothesized that hormonal changes, starvation, and increased risk to develop infection might jeopardize the outcome of grafts from donors with a prolonged ICU stay. Although 70% of the early hepatocellular injuries are reversible, the remaining 30% result in graft failure.

Percutaneous gastrostomy and gastroenterostomy: 1. Techniques derived from laboratory evaluation

Various techniques, guidance systems, instruments, and the postmortem effects of percutaneous gastrostomy (PG) and percutaneous gastroenterostomy (PGE) were evaluated in 30 laboratory animals and five human cadavers. Methods to distend the stomach included air, fluid, intragastric balloon, and percutaneous needle inflation; a variety of trocar systems and catheters inserted by Seldinger technique (including those adapted from other uses and several designed specifically) were assessed. Fluoroscopy was the preferred guidance system, though sonography proved valuable (liver position, depth calculation to the stomach, localization of vessels to avoid), and the entire PG procedure was performed under sonographic guidance in four animals. Although the procedure was safe in most cases, several major complications did occur: laceration of a low-lying liver with exsanguination, malpositioned catheters in the lesser sac and adjacent to the spleen, and violation of the backwall of the stomach with laceration of celiac and splenic vessels. The animals and cadavers underwent autopsy. Autopsy revealed that firm gastrocutaneous tracts were formed by 7 days. There were few instances of wound infection, intraperitoneal fluid leakage, or evidence of trauma to the stomach when the catheters were well seated. Injury to the inferior epigastric artery is a potential hazard, and in cadaver dissections was located between the middle third and outer margin of the rectus abdominis muscle. Laboratory experience has been, and continues to be, an important means to improve and use new techniques for PG and PGE.