Diagnostic laparoscopy and laparoscopic ultrasonography optimize the staging and resectability of intraabdominal neoplasms

BACKGROUND

Despite technical improvements, preoperative imaging studies often fail to predict intraoperative findings. We investigated the potential use of diagnostic laparoscopy (DL) and laparoscopic ultrasonography (LUS) for the assessment of disease in patients with abdominal neoplasms.

METHODS

Fifty consecutive patients with abdominal neoplasms underwent spiral computed tomography with oral and intravenous contrast using 5-mm contiguous sections. In addition, eight patients underwent ultrasonography, six underwent magnetic resonance imaging, and eight underwent positron emission tomography. All patients then underwent DL and LUS using a 7.5-MHz ultrasound probe.

RESULTS

There were 29 men and 21 women with a mean age of 63 years (range, 35-84). Most had a diagnosis of colorectal cancer (19 cases), melanoma (12 cases), or hepatoma (five cases). In nine cases (18%), DL revealed peritoneal metastatic implants not shown on preoperative images. In 18 cases (36%), LUS was more accurate than preoperative imaging. Combined DL and LUS findings radically changed the operative management in 16 patients (32%).

CONCLUSION

As compared with preoperative imaging, the combination of DL and LUS provides more accurate information regarding staging and resectability. Moreover, it helps to determine the extent of operation and reduces the number of unnecessary laparotomies. DL and LUS should be used as an adjunct to preoperative imaging studies in patients with primary or metastatic intraabdominal neoplasms.

Surgical resection for metastatic melanoma to the liver: the John Wayne Cancer Institute and Sydney Melanoma Unit experience

HYPOTHESIS

Metastatic melanoma to the liver is not incurable; complete surgical resection can achieve long-term survival in selected patients.

BACKGROUND

Metastases to the liver are diagnosed in 10% to 20% of patients with American Joint Committee on Cancer stage IV melanoma. Surgical resection has not been generally accepted as a therapeutic option, as most patients will have other sites of disease that limit their survival to a median of only 4 to 6 months. However, there is little information on outcomes following resection in those patients with disease limited to the liver.

PATIENTS AND METHODS

Review of the prospective melanoma databases at the John Wayne Cancer Institute, Santa Monica, Calif, and the Sydney Melanoma Unit, Sydney, Australia, identified 1750 patients with hepatic metastases, of whom 34 (2%) underwent exploration with intent to resect the metastases. Prognostic factors within the group of patients who underwent resection were examined by univariate and multivariate analysis, and median disease-free survival (DFS) and overall survival (OS) were calculated.

RESULTS

Of 34 patients undergoing exploratory celiotomy, 24 (71%) underwent hepatic resection and 10 (29%) underwent exploration but not resection. Eighteen patients (75%) underwent complete surgical resection, while the remaining 6 underwent palliative or debulking procedures with incomplete resection. The operative resections included lobectomy (n=14), segmentectomy (4), nonanatomic resection (5), and extended lobectomy (1). The median number of resected lesions was 1, and median lesion size was 5 cm (range, 0.7-22 cm). The median disease-free interval between initial diagnosis of melanoma and development of hepatic metastases was 58 months (range, 0-264 months). Median DFS and OS estimates in the 24 patients who underwent surgical resection were 12 months (range, 0-147 months) and 28 months (range, 2-147 months), respectively. Five-year DFS and OS in this group were 12% and 29%. Macroscopically, complete resection of disease (P =.001) and histologically negative resection margins (P =.03) significantly improved DFS by univariate analysis. Patients rendered surgically free of disease also tended to have improved OS (P =.06). Median OS was 28 months for patients who underwent surgical resection compared with 4 months for patients who underwent exploration only (P<.001).

CONCLUSIONS

Resection of metastatic melanoma to the liver may improve DFS and OS in selected patients, similar to resection of other metastatic sites. Therefore, patients with limited metastatic sites, including the liver, who can be rendered free of disease should be considered for complete surgical resection, as their prognosis is otherwise dismal.

Interleukin-2 binds to ganglioside GD(1b)

We have developed a solid matrix immunoassay to determine the binding of interleukin-2 (IL-2) to specific gangliosides. The assay establishes that recombinant human IL-2 binds to ganglioside GD(1b) but not to any other gangliosides (GM(1), GM(2), GM(3), GD(1a), GD(2), GD(3), and GT(1b)). The binding varies with the ratio of GD1b and IL-2. This assay enables distinguishing the nature of the sugar moiety of the ganglioside recognized by IL-2 and establishes the dosimetry of the ganglioside-IL-2 interaction. Since rIL-2 is administered systematically into stage IV melanoma patients, we have examined 45 tumor biopsies for GD(1b) content. The incidence of GD(1b) in tumor biopsies is 51%. We postulate that GD(1b) associated on the tumor or in the circulation of cancer patients may bind to rIL-2 and prevent the availability of rIL-2 to augment antitumor-immune response.

Molecular strategy for detecting metastatic cancers with use of multiple tumor-specific MAGE-A genes

BACKGROUND

The human melanoma-associated antigen family A (MAGE-A) has high specificity and expression in various malignancies, but individual family members are expressed at low frequency in any one particular type of cancer. We therefore developed a method to detect mRNAs from multiple MAGE-A genes in a single reaction.

METHODS

Universal MAGE-A (uMAGE-A) primers and probe were designed to reverse-transcribe, amplify, and detect by electrochemiluminescence (ECL) MAGE-A mRNAs on the Origen Analyzer. The assay was performed on total RNA of melanoma (n = 9 cell lines and 24 tumors), breast cancer (n = 7 and 26), and colorectal cancer (CRC; n = 5 and 12). We also evaluated blood from melanoma (n = 50), breast cancer (n = 16), and CRC (n = 21) patients.

RESULTS

The uMAGE-A mRNA was detectable in 0.01-1 ng of cell line RNA. The identity of the uMAGE-A cDNA products was confirmed by sequencing and polyacrylamide gel electrophoresis. The uMAGE-A assay increased detection of melanoma, breast cancer, and CRC tumor by 13%, 31%, and 25%, respectively, compared with a MAGE-A1 assay, and by 17%, 19%, and 25%, respectively, compared with a MAGE-A3 assay. The uMAGE-A assay detected circulating tumor cells in the blood of melanoma (24%), breast cancer (25%), and CRC (29%) patients.

CONCLUSIONS

The uMAGE-A reverse transcription-PCR/ECL assay provides a practical and sensitive approach for detection of various metastatic cancers in tissues and blood.

Does human melanoma express carcinoembryonic antigen?

BACKGROUND

Several investigators have proposed that carcinoembryonic antigen (CEA), an immunogenic antigen expressed by colon carcinoma, may also be expressed by human melanoma. Because sialyl Lewisx (sLex), the carbohydrate moiety of CEA, has been identified in melanoma, we compared CEA and sLex levels in colon carcinoma cells and melanoma cells.

METHODS

CEA levels were assessed for expression on the cell surface and in cell lysates of cutaneous melanoma cell lines by two different kinds of ELISA, and by Western blot analysis of immunoprecipitated CEA using monoclonal antibodies (Mabs) T84-66 and COL-1, which have defined specificities for CEA. Colon carcinoma cells and purified CEA were positive controls.

RESULTS

Both Mabs reacted strongly with cell surface and cell lysates of colon cancer. Mab T84-66 reacted well with cell surface but not cell lysates of melanoma. COL-1 reacted poorly with cell surface but its binding increased with the density of melanoma cell lysates. Both Mabs intensely stained the blots of purified CEA and colon carcinoma lysates immunoprecipitated with the respective Mabs, but failed to stain the immunoprecipitates of melanoma cell lysates. Both Mabs bound to lysates immunoprecipitated with anti-sLex Mab in colon carcinoma, but not in melanoma. Cell-surface expression of CEA and sLex was significantly correlated (r2: 0.88) in colon cancer cells but not in melanoma.

CONCLUSION

Our results confirm the presence of CEA in colon carcinoma but not in human cutaneous melanoma cell lines.

Gangliosides as targets for immunotherapy for pancreatic adenocarcinoma

BACKGROUND

Pancreatic adenocarcinoma cells express gangliosides and sialyl Lewis (sLe) antigens. It is not known whether these carbohydrate antigens can be targeted by immunotherapy. The authors measured the expression of GM(2) and sLe antigens on the surface of pancreatic carcinoma cells and the serum levels of total gangliosides, GM(2), and antiganglioside antibodies in patients with pancreatic carcinoma.

METHODS

Cell surface GM(2) and sLe antigens were measured by cell suspension enzyme linked immunoadsorbent assay (ELISA) in four pancreatic carcinoma cell lines. Sera from 20 pancreatic carcinoma patients and 20 age- and gender-matched healthy volunteers were analyzed for antiganglioside and anti-sLe immunoglobulin (Ig) M titers by ELISA. Serum levels of total gangliosides and GM(2) also were measured.

RESULTS

All cell lines expressed GM(2) and sLe antigens. When compared with age- and gender-matched volunteers, patients had significantly higher serum levels of total gangliosides (25.6 +/- 9.0 mg/dL vs. 15.6 +/- 2.7 mg/dL; P < 0.001), GM(2) (0.278 +/- 0.415 mg/dL vs. 0.013 +/- 0.018 mg/dL; P = 0.02), ELISA units of anti-GM(2) IgM antibody (368 +/- 95 vs. 155 +/- 25; P = 0.04) and anti-GD(1b) IgM antibody (351 +/- 91 vs. 138 +/- 26; P = 0.03), but not anti-sLe(x) IgM (1389 +/- 345 vs. 1081 +/- 224; P = 0.46) or anti-sLe(a) IgM antibody (1097 +/- 253 vs. 1200 +/- 315; P = 0.80). Patients with unresectable tumors had higher serum levels of total gangliosides compared with patients with resectable tumors, and a serum level > 25 mg/dL was found to correlate significantly with poor overall survival (P < 0.02).

CONCLUSIONS

Increased serum levels of total gangliosides and GM(2) may reflect shedding or release of gangliosides from the surface of tumor cells. Production of IgM antibody against GM(2) and GD(1b) indicates that these gangliosides are immunogenic antigens that may be potential targets for effective active immunotherapy.

Molecular detection of metastatic pancreatic carcinoma cells using a multimarker reverse transcriptase-polymerase chain reaction assay

BACKGROUND

The diagnosis of pancreatic carcinoma is often associated with a poor prognosis, because most patients already have advanced disease. A highly sensitive assay to detect the progression of pancreatic carcinoma would be of significant clinical utility. The authors developed multiple tumor mRNA markers for reverse transcriptase-polymerase chain reaction (RT-PCR) to detect metastatic tumor cells in the blood and tissue of patients with American Joint Committee on Cancer (AJCC) Stage II/III or IV pancreatic carcinoma.

METHODS

An RT-PCR plus Southern blot assay was used to detect mRNA of tumor markers in blood and tissues. mRNA expression of the tumor progression markers MET (hepatocyte growth factor receptor gene c-met), GalNAc-T (beta1,4- N-acetyl-galactosaminyl-transferase), and beta-hCG (beta-human chorionic gonadotropin) was evaluated in 9 pancreatic carcinoma cell lines, 13 tumor biopsy specimens, 5 nonmalignant pancreatic tissue specimens, and blood from 33 pancreatic carcinoma patients and 32 healthy donors.

RESULTS

The detection limit of the assay was 1 rhog, 10 rhog, and 10 rhog for MET, GalNAc-T, and beta-hCG mRNA expression, respectively. The pancreatic carcinoma cell lines expressed all three mRNA markers. Of blood specimens from 17 patients with AJCC Stage IV pancreatic carcinoma, 82%, 65%, and 76% were MET, GalNAc-T, and beta-hCG mRNA positive, respectively. Of blood specimens from 16 patients with AJCC Stage II/III disease, 88% were positive for at least 1 mRNA marker.

CONCLUSIONS

A multiple molecular marker assay was developed to detect cancer cells in blood and tissue from patients with different stages of pancreatic carcinoma. The detection of cancer cells in the blood may be used as a marker of pancreatic tumor progression and may be useful in monitoring response to therapy.

Molecular detection of metastatic pancreatic carcinoma cells using a multimarker reverse transcriptase-polymerase chain reaction assay

BACKGROUND

The diagnosis of pancreatic carcinoma is often associated with a poor prognosis, because most patients already have advanced disease. A highly sensitive assay to detect the progression of pancreatic carcinoma would be of significant clinical utility. The authors developed multiple tumor mRNA markers for reverse transcriptase-polymerase chain reaction (RT-PCR) to detect metastatic tumor cells in the blood and tissue of patients with American Joint Committee on Cancer (AJCC) Stage II/III or IV pancreatic carcinoma.

METHODS

An RT-PCR plus Southern blot assay was used to detect mRNA of tumor markers in blood and tissues. mRNA expression of the tumor progression markers MET (hepatocyte growth factor receptor gene c-met), GalNAc-T (beta1,4- N-acetyl-galactosaminyl-transferase), and beta-hCG (beta-human chorionic gonadotropin) was evaluated in 9 pancreatic carcinoma cell lines, 13 tumor biopsy specimens, 5 nonmalignant pancreatic tissue specimens, and blood from 33 pancreatic carcinoma patients and 32 healthy donors.

RESULTS

The detection limit of the assay was 1 rhog, 10 rhog, and 10 rhog for MET, GalNAc-T, and beta-hCG mRNA expression, respectively. The pancreatic carcinoma cell lines expressed all three mRNA markers. Of blood specimens from 17 patients with AJCC Stage IV pancreatic carcinoma, 82%, 65%, and 76% were MET, GalNAc-T, and beta-hCG mRNA positive, respectively. Of blood specimens from 16 patients with AJCC Stage II/III disease, 88% were positive for at least 1 mRNA marker.

CONCLUSIONS

A multiple molecular marker assay was developed to detect cancer cells in blood and tissue from patients with different stages of pancreatic carcinoma. The detection of cancer cells in the blood may be used as a marker of pancreatic tumor progression and may be useful in monitoring response to therapy.

Do all patients with sentinel node metastasis from breast carcinoma need complete axillary node dissection?

OBJECTIVE

To determine the likelihood of nonsentinel axillary metastasis in the presence of sentinel node metastasis from a primary breast carcinoma.

SUMMARY BACKGROUND DATA

Sentinel lymphadenectomy is a highly accurate technique for identifying axillary metastasis from a primary breast carcinoma. Our group has shown that nonsentinel axillary lymph nodes are unlikely to contain tumor cells if the axillary sentinel node is tumor-free, but as yet no study has examined the risk of nonsentinel nodal involvement when the sentinel node contains tumor cells.

METHODS

Between 1991 and 1997, axillary lymphadenectomy was performed in 157 women with a tumor-involved sentinel node. Fifty-three axillae (33.5%) had at least one tumor-involved nonsentinel node. The authors analyzed the incidence of nonsentinel node involvement according to clinical and tumor characteristics.

RESULTS

Only two variables had a significant impact on the likelihood of nonsentinel node metastasis: the size of the sentinel node metastasis and the size of the primary tumor. The rate of nonsentinel node involvement was 7% when the sentinel node had a micrometastasis (< or =2 mm), compared with 55% when the sentinel node had a macrometastasis (>2 mm). In addition, the rate of nonsentinel node tumor involvement increased with the size of the primary tumor.

CONCLUSIONS

If a primary breast tumor is small and if sentinel node involvement is micrometastatic, then tumor cells are unlikely to be found in other axillary lymph nodes. This suggests that axillary lymph node dissection may not be necessary in patients with sentinel node micrometastases from T1/T2 lesions, or in patients with sentinel node metastases from T1a lesions.

Complications of hepatic cryosurgery

Cryosurgery may be considered for patients whose hepatic lesions are not amenable to surgical resection, i.e., patients with multiple hepatic lesions and/or lesions abutting major vascular structures. Because the size of the iceball created during the procedure can be carefully controlled, cryosurgery has the advantage of being a focal technique that spares much more noncancerous liver tissue than surgical resection. The major complications of hepatic cryosurgery are the same as those of hepatic resection: hemorrhage, pleural effusion, bile leak fistula, perihepatic abscess, and hepatic failure. In addition, there is a risk of coagulopathy when large tumors are frozen using multiple freeze-thaw cycles. In general, operative morbidity is related to the volume of frozen tissue, the number of freeze-thaw cycles, and number of cryoprobes. Further experience and accrual of long-term data should better define the indications for hepatic cryosurgery and minimize the incidence of complications.

Fine needle aspiration cytology of mucinous cystadenocarcinoma of the pancreas. Further observations

Mucinous cystadenocarcinoma of the pancreas is among the few potentially curable malignancies of this organ; recognition of the differences between this and other pancreatic tumors is thus of paramount importance. A second example of mucinous cystadenocarcinoma of the pancreas diagnosed by fine needle aspiration, performed under ultrasonographic guidance, is presented, emphasizing both the radiographic distinction from other cystic lesions plus cytologic features not previously stressed. In essence, the cytodiagnosis relies less on cell morphology, which is characteristically bland and relatively monomorphic, than on the presence of extracellular mucin, which must be distinguished from amorphous background material and necrotic debris. The possibility of diagnosing benign mucinous and serous cystadenomas by imaging and cytology is addressed.