Tumori Journal

To evaluate the cost effectiveness of screening by colposcopy compared to conventional cytology.

A prospective study was performed in the Florence District screening center on 3,000 consecutive women, self referring, who were examined by cytology and colposcopy in a blind fashion. Further assessment was based on cytologic report or on colposcopy-directed punch biopsy. Actual costs of the whole screening process were known. The cost effectiveness of different possible simulated screening scenarios was then determined.

Overall, 18 high-grade lesions (CIN3 = 9, CIN2 = 9) were detected. Four different screening scenarios were compared, namely a) cytology alone, b) cytology + repeat smear for ASCUS (atypical squamous cells of undeterminate significance) cases, c) colposcopy + cytology for cases of condyloma at punch biopsy, and d) colposcopy alone. Although they had a higher cost per examined woman (a) = 17.98, b) = 19.40, c) = 23.86, d)= 22.10 US$), scenarios c) and d) had a higher relative sensitivity (a = 44.4, b = 61.1, c = 100, d = 88.8%) and a lower cost per high-grade lesion detected (a = 6,743, b = 5,291, c = 3,977, d = 4,144 US$).

Screening by colposcopy is a feasible procedure which is more sensitive and more cost effective than conventional cytologic screening. At least in those settings where access to cytopathology may be difficult, screening by colposcopy should be considered as a possible alternative.

Melanoma of the external ear is a rare disease, and its management is controversial. To address this problem, we reviewed the data concerning the patients observed at our Institution.

We retrospectively reviewed the clinical records of the 20 patients bearing primary ear melanoma observed over a period of about 20 years at the Istituto Nazionale Tumori of Milan.

Initial evaluation of the patients revealed 7 stage I, 12 stage II and 1 stage III. The thickness of the tumors varied from 0.39 to 6.62 mm. Fourteen patients underwent a wedge resection of the skin and cartilage with primary closure, and 6 patients had a partial amputation of the ear. In 8 cases the section was performed at about 1 cm from the border of the tumor, in 6 cases at about 0.5 cm, and in 6 cases at more than 1 cm. The average follow-up was 57 months (range, 1-18 years). Since there was no local recurrence, it could not be related to type and extent of the local resection performed. In contrast, the development of metastases was related to tumor thickness.

A conservative excision with margins of 1 cm can be a safe procedure for invasive ear melanoma, irrespective of tumor thickness. Like melanomas of other sites, the prognosis is linked to the thickness of the tumor.

Several studies have emphasized the role of radiation therapy for patients with pelvic recurrences of rectal carcinoma following primary surgery. The occurrence of local-regional relapse usually means a poor prognosis and often a poor quality of life, so that different authors consider the prognosis of patients relapsing after surgery worse than those with primary inoperable tumors or those with residual disease after resection.

Between January 1988 and January 1995, 43 patients with local recurrence of rectal carcinoma were treated at our Institution. Twenty-three had previously been operated by abdominoperineal resection and 20 by anterior resection. Thirteen cases also received adjuvant chemotherapy. All patients underwent irradiation with a 6-15 MeV linear accelerator; 8 (19%) received a total dose of up to 45 Gy on the pelvis and 35 (81%) higher than 45 Gy. Eighteen cases (42%) underwent 3-6 courses of chemotherapy with 5-fluorouracil and folates during radiation.

Treatment tolerance was satisfactory. All cases underwent restaging at 45 days from completion of treatment. Sixteen cases (37%) showed a radiologic response >50%. Median overall survival after relapse was 18.8 months. There were no statistical significant differences in survival between patients treated exclusively with radiation and those treated with chemo-radiothera-py (17 vs 22 months). The results of patients who received doses higher than 45 Gy were statically better (P < 0.05) than those irradiated up to 45 Gy. A slight increase in survival was demonstrated in cases submitted to radical surgery after combined treatment (25 months). Twenty-seven cases (63%) obtained pain control after radiation therapy (median pain remission, 11 months).

Our results seem to encourage radiation therapists, surgeons and oncologists to have a more curative attitude in the treatment of selected patients with local-regional recurrences of rectal cancers by using multi-modality therapy.

The orbit is an uncommon primary site for non-Hodgkin's lymphomas (NHL), and it accounts for less than 1% of all sites of primary presentations. We report the experience of the Department of Radiation Oncology at Ankara University Faculty of Medicine with radiation therapy in treatment of patients with stage I primary orbital NHL.

From February 1978 through August 1993, 14 patients with stage I primary orbital NHL were treated with radiation therapy. According to the Working Formulation classification, 8 patients had low-grade and 6 had intermediate-grade lymphomas. The most commonly used radiation therapy technique was a single anterior field with a Cobalt-60 unit, delivering 40 Gy in 2 Gy daily fractions. Two patients with intermediate-grade lymphomas received the CHOP regimen following radiation therapy.

Follow-up ranged from 0.8 to 18.3 years (median, 10.3 years). Local control was achieved in all patients. Two patients with low-grade lymphomas relapsed locally and were successfully salvaged with radiation therapy. Three patients with intermediate-grade lymphomas failed systemically. Salvage therapy consisted of combination chemotherapy for 2 of them but was unsuccessful. Overall survival probabilities at 2, 5 and 10 years were 78.6%, 61.1% and 52.4%, respectively, for the entire group of 14 patients. Overall, cause-specific and disease-free survival probabilities were higher for patients with low-grade lymphomas than for those with intermediate-grade lymphomas (P = 0.03, P = 0.03 and P = 0.06, respectively). Cataracts were observed in 9 and lacrimal disorders in 4 patients.

The study suggests that among stage I primary orbital NHL, low-grade lymphomas could be treated with radiation therapy alone, whereas combination chemotherapy could accompany radiation therapy for intermediate-grade lymphomas.

Vinorelbine, a new semi-synthetic vinca alkaloid, has demonstrated high activity as a single agent in pretreated metastatic breast cancer.

To evaluate the activity of the combination vinorelbine-mitomycin C and to reduce the incidence of side effects, in particular myelotoxicity, patients with metastatic breast cancer pretreated with 1 or more chemotherapy regimens for metastatic disease were treated according to the following schedule: mitomycin C, 10 mg/m² day 1, and vinorelbine, 25 mg/m², days 1, 8 and 15, every 28 days.

Twenty-seven patients were enrolled and were evaluable for activity and side effects. A total of 157 cycles were delivered (median 5 cycles per patient). There were 10 partial remissions (37%; 95% confidence interval 20-59%), 5 instances of stable disease and 12 of disease progression. Grade III-IV toxicity was mostly hematological and included thrombocytopenia (4%) and neutropenia (42%).

Our results indicate that the combination of mitomycin C and vinorelbine has moderate activity in pretreated breast cancer.

Ovarian carcinoma remains confined to the peritoneal cavity for the greater part of its natural history, so intraperitoneal (IP) administration of chemotherapy could result in greater total drug exposure of the tumor and minimize systemic antiblastic drug side effects. The aim of this study was to evaluate the therapeutic efficacy and toxic effects of intraperitoneal mitoxanthrone in patients affected by ovarian carcinoma with macroscopic absence of disease or minimal residual disease.

Ten patients were enrolled (stage II and III) who had been previously treated with neoadjuvant systemic chemotherapy (CDDP or CBDCA + CTX) and radical surgery resulting in macroscopic absence of disease or minimal residual disease (<1 cm). Mitoxanthrone (25 mg/m²) was instilled in 2 liters of normal saline every four weeks for 2-4 cycles.

A total of 26 courses was administered; two patients discontinued IP therapy, one for chemoperitonitis and another for bowel perforation requiring catheter removal. Of the 10 patients receiving IP chemotherapy, 7 are alive at 5 years from radical surgery, and 3 had relapses at 13, 14 and 57 months, respectively, from radical surgery.

Intraperitoneal mitoxanthrone appears to be an effective second-line therapy in ovarian cancer; it is well tolerated as far as toxic effects are concerned, allowing cost reduction and improved patient compliance. For those cases requiring a limited number of peritoneal accesses traditional percutaneous systems have a more favorable cost/benefit ratio.

Giant cell tumor of tendon sheath (GCTTS) is a common tumor occurring on the tendon sheaths of the fingers. The nature of this lesion is still controversial: some researchers consider it a reactive process arising from chronic inflammation while others regard it as a tumor of presumed synovial cell or monocytic macrophage system origin. In an effort to clarify the histogenesis we decided to further investigate the immunophenotypic profile of this tumor.

We studied 20 GCTTS of the fingers using a panel of 18 antibodies, 13 monoclonal and 5 polyclonal.

The immunohistochemical investigation revealed that the mononuclear cells of this lesion can be divided into two groups. The cells of the first and more numerous group were positive for vimentin, PG-M1 and KP1 but also for muscle actin (HHF35 monoclonal antibody) and neuron-specific enolase. A second population of mononuclear cells, usually arranged around the giant cells, were positive for PG-M1, KP1, LCA and occasionally for alpha-1-antitrypsin and alpha-1-antichymotrypsin. Multinucleated giant cells were also positive for KP1, PG-M1 and LCA monoclonal antibodies. A variable but usually weak positivity for al-pha-1-antitrypsin, alpha-1-antichymotrypsin and lysozyme was also observed.

Our results suggest a synovial cell origin for GCTTS and do not support the hypothesis that it could be a neoplasm with a true histiocytic origin. The positivity of some cells for the HHF35 antibody, together with electron microscopic evidence of filament bundles with focal dense bodies, suggests that at least part of the mononuclear cells may have a myofibroblastic differentiation.

Structurally altered proteins (derived from chromosomal translocations or gene mutations) can be considered tumor specific antigens and represent an attractive target for a T-cell mediated response. T lymphocytes recognize antigens in the form of peptides bound to HLA-mole-cules.

Peptides derived from oncogenic proteins were screened fro the presence of HLA binding motifs; actual binding were evaluated by HLA stabilization experiments using transfectants and specific anti-HLA antibodies. Specific lymphocytes were induced by in vitro peptide sensitization and screened by thymidine uptake or cellular cytotoxic assays.

We identified peptides derived from EWS/FLI-1 fusion protein and from mutated K-RAS protein (encompassing respectively the fusion point and the mutation at position 12) that showed binding motif for HLA-Cw^*0702 and HLA-A3 respectively. The actual binding of these peptides was analysed in a stabilization assay. We detected binding for the EWS/FLI-I peptide and for 5 RAS peptides (1 wild type and 4 mutated). The effect of temperature, β2-microglobulin (β2-m) and fetal calf serum (FCS) on the binding and the stability of the HLA/peptide complex was studied. A low temperature (26°C) increased the binding both in HLA-A3 and HLA-Cw^*0702, while FCS reduced it. β2-m increased the binding to the HLA-A3 molecule but did not influence the binding to the HLA-Cw^*0702. The stability of already formed complexed was somewhat different in the HLA-A3 and HLA-Cw^*0702 system: both were more stable at 26°C than at 37°C but while the β2-m and FCS did not influence the stability of the HLA-A3/peptide complex, they seemed to cause opposite effects in the HLA-Cw^*0702 system (β2-m stabilized and FCS destabilized the complex). Finally, we were able to generate a specific CD8⁺ CTL line against a K-RAS mutated peptide.

Although binding motifs and actual HLA binding can be detected in several cases, the generation of a cellular response is infrequent, confirming that HLA binding is necessary but not sufficient to obtain an in vitro response. Further optimization of culture conditions, type of Antigen Presenting Cells (APC), peptides, use of stabilizers like β2-m are still needed.