Outcomes and factors impacting local recurrence of ductal carcinoma in situ.

BACKGROUND: The optimal management of ductal carcinoma in situ (DCIS) remains controversial. Investigators have focused on identifying patients who are eligible for treatment by excision alone. A retrospective analysis of patients with DCIS treated by various modalities was conducted to compare outcomes and determine factors significant for local recurrence (LR). METHODS: Between 1985-1992, 88 consecutive diagnoses of DCIS were identified in 85 patients. Seventy-four percent were detected mammographically. The most common histologic subtypes were comedo (54%) and cribriform (23%). Tumor sizes were < 2.5 cm (49%), > 2.5-5 cm (26%), > 5 cm (23%), and unknown (2%). Final resection margins were tumor free (75%), close/positive (23%), and unknown (2%). Treatment methods included mastectomy (30%), localized surgery and radiation therapy (LSR) (43%), or wide localized surgery alone (LS) (27%). Radiation therapy (RT) was comprised of 50 grays to the breast, and 53% of treated patients received local "boost" irradiation. RESULTS: The median follow up was 8.3 years. The overall recurrence rate was 13. 6%, whereas the median time to LR was 27.8 months. Recurrence rates according to treatment modality were: LS: 25%; LSR: 13%; and mastectomy: 4%. However, if surgical margins were tumor free, LSR had a LR rate of 3.4%. After RT, no LR occurred prior to 15 months, and 4 of 5 tumors were noninvasive. Nine patients treated by excision alone conformed to the criteria of Lagios et al. criteria and LR occurred in three of nine tumors. Of the factors analyzed, margin status was found to be the best predictor for LR (P = 0.05). CONCLUSIONS: If surgical margins are tumor free, the LSR regimen is equivalent to mastectomy for local tumor control. Annual mammograms may be adequate for the follow-up of patients with irradiated breasts, but biannual studies still are recommended for patients treated with excision alone.

The role of adjuvant radiotherapy in the treatment of resectable desmoid tumors.

PURPOSE: Desmoid tumors have a high propensity for local recurrence with surgical resection. There are many reports describing good responses of desmoid tumors to irradiation, but none have clearly established the indications for adjuvant radiotherapy in treating resectable desmoid tumors. METHODS AND MATERIALS: A retrospective analysis was performed on 61 patients with resectable desmoid tumor(s) who were treated at our institution from 1965 to February of 1992. Five patients had multifocal disease and are analyzed separately. Fifty-six patients had unifocal disease, of which 34 had positive surgical margins. Forty-five of the 56 patients with unifocal disease were treated with surgery alone, while 11 were treated with surgery plus adjuvant radiotherapy. Median follow-up was 6 years. Local control was measured from the last day of treatment, and all cases were reviewed by our Department of Pathology. RESULTS: Multivariate analysis of the 56 patients with unifocal disease revealed that positive margins independently predicted for local recurrence (p < or = 0.01). Only 3 of 22 patients with clear margins experienced a local recurrence, with a 6-year actuarial local control of 85%. Multivariate analysis of the 34 patients with positive margins revealed that adjuvant radiotherapy independently predicted for improved local control (p = 0.01), and patients with recurrent disease had a slightly higher risk of local recurrence (p = 0.08). The 6-year actuarial local control determined by Kaplan-Meier for patients with unifocal disease and positive margins was 32% (+/-12%) with surgery alone, and 78% (+/-14%) with surgery plus adjuvant radiotherapy (p = 0.02). Subgroup analysis of the patients with positive margins and recurrent disease revealed that those treated with surgery alone had a 6-year actuarial local control of 0% vs. 80% for those treated with surgery plus radiotherapy (p < or = 0.01). Patients with positive margins and primary disease had a trend towards improved local control with adjuvant radiotherapy, but this was not statistically significant. None of the patients treated with radiotherapy developed serious complications or a secondary malignancy. CONCLUSIONS: Margin status is the most important predictor of local recurrence for patients with resectable, unifocal desmoid tumor. Adjuvant radiotherapy is indicated in the treatment of patients with positive margins following wide excision of recurrent disease. The role of adjuvant radiotherapy in patients with positive margins following resection of primary disease is controversial, and should be based on a balanced discussion of the potential morbidity from radiotherapy compared to the potential morbidity of another local recurrence. Adjuvant radiotherapy is less likely to benefit those with clear margins due to the excellent results for these patients treated with surgery alone. The local control of desmoid tumor in the adjuvant setting is excellent with total doses ranging from 50-60 Gy, with acceptable morbidity. Field sizes should be generous to prevent marginal recurrences, and large volume MRIs of patients with extremity lesions should be used to identify those patients with multifocal disease.

Radiation recall dermatitis and pneumonitis in a patient treated with paclitaxel.

BACKGROUND: Radiation recall refers to a tissue reaction produced by a chemotherapeutic agent in a previously irradiated field that would not occur in a nonirradiated field. A number of agents have been reported to cause radiation recall. Recently, there have been case reports of recall dermatitis from paclitaxel treatment. METHODS: A patient with metastatic lung cancer received palliative radiation to her mediastinum and ribs. Because of disease progression, she subsequently received paclitaxel. RESULTS: After paclitaxel administration, the patient became acutely dyspneic. A subsequent chest X-ray revealed a parenchymal opacity in a region that corresponded with the patient's radiation portal. She also developed a severe skin reaction in the previously treated electron field. CONCLUSIONS: This is one of few reported cases of recall dermatitis from paclitaxel and is also suggestive of recall pneumonitis, a phenomenon previously unreported to the authors' knowledge. Given paclitaxel's ability to function as a radiosensitizer, this response is not unexpected. As the frequency of paclitaxel administration increases, its potential as a radiation sensitizer and radiation recall should be considered.

Presence of a specific antiestrogen binding site on human follicular thyroid carcinoma cell line (UCLA RO 82 W-1): inhibition by an endogenous ligand present in human serum.

A receptor for antiestrogens, distinct from the estrogen receptor, has been identified in several tissues including the MCF-7 breast cancer cell line. Estrogen receptors have also been found in normal and pathological thyroid tissue homogenates. We demonstrate the presence of an antiestrogen binding site (AEBS) on a pure human follicular thyroid carcinoma cell line (UCLA RO 82 W-1) using a 3H-tamoxifen (3H-TAM) binding assay. The binding of 3H-TAM to the AEBS was determined after preincubation (30 min) of the cells with excess 17 beta-estradiol (2 mumol/L). Specific and saturable binding of 3H-TAM to the cells was observed. Displacement of the tracer from its binding site was dose dependent. Scatchard analysis revealed a dissociation constant (Kd) of 73 nmol/L, indicating a binding site with moderate affinity and capacity (72 pmol/10(6) cells). Using this assay we were also able to demonstrate the presence of an endogenous ligand for the AEBS in ethanol extracts of human serum. Cell growth and 3H-thymidine incorporation by the follicular thyroid carcinoma cells were inhibited when the cells were exposed to TAM (1.5 mumol/L). In conclusion, TAM is able to bind to a specific receptor on this follicular thyroid carcinoma cell line, and a natural circulating ligand present in ethanol extracts of human serum interferes with its binding.

Tumor necrosis factor-alpha activates nuclear factor kappa B and induces manganous superoxide dismutase and phosphodiesterase mRNA in human papillary thyroid carcinoma cells.

Human papillary thyroid carcinoma (PTC) has a relatively benign prognosis despite a high frequency of lymphatic metastasis. This suggests that local anticancer factors, generated in lymph nodes, control PTC progression. The cytokine, tumor necrosis factor-alpha (TNF-alpha), may be one such factor. We have previously shown that a human PTC cell line (NP-PTC) has high affinity TNF-alpha receptors. We now report on the action of TNF-alpha in these cells. TNF-alpha decreased [3H]thymidine incorporation as well as cellular DNA content and cell number in a dose-dependent manner. The abundance of phosphodiesterase and manganous superoxide dismutase mRNA species was increased in a time- and dose-dependent manner in the NP-PTC cells after TNF-alpha treatment. TNF-alpha activated NF-kappa B, a nuclear factor thought to mediate multiple actions of TNF-alpha, in these cells with a maximum effect observed after 30 min of treatment. Thus, TNF-alpha has an antiproliferative action on NP-PTC cells, despite its ability to induce the accumulation of mRNA that encodes an enzyme (manganous superoxide dismutase), thought to be cytoprotective. The net antiproliferative effect must therefore be explained by a balance of protective and tumoricidal or static effects that ultimately result in control of tumor spread. These antiproliferative effects may be in part mediated by NF-kappa B and PDE.

Stages 1 and 2 epidermoid carcinoma of the glottic larynx: involvement of the anterior commissure.

Data on 55 patients with stages 1 and 2 epidermoid carcinoma of the glottic larynx treated from 1978 to 1988 were retrospectively reviewed. Twenty-six patients had involvement of the anterior commissure (AC). Local and ultimate local control rates achieved, respectively, with mean follow-up of 41 months (range, 6-120 months), were as follows: 92% and 100% for patients without AC involvement and stage T1a lesions, 60% and 80% for patients without AC involvement and stage T2 lesions, 100% and 100% for patients with AC involvement and stage T1a lesions, 100% and 100% for patients with AC involvement and stage T1b lesions, and 75% and 100% for patients with AC involvement and stage T2 lesions. There was no correlation between the degree of response at completion of treatment and local control. There was no difference in the local control rate of patients with and without involvement of the AC. Factors associated with a decreased local control rate include extensive subglottic extension and use of a single lateral field technique. Surgical salvage after failure of radiation therapy is effective and can be performed with acceptable morbidity.

Concurrent cisplatin and 5-fluorouracil chemotherapy and radiation therapy for advanced-stage squamous carcinoma of the vulva.

A phase II trial of concurrent cisplatin and 5-fluorouracil (5-FU) chemotherapy and radiation therapy (CT + RT) was conducted for the primary treatment of 12 patients with retrospective surgical FIGO stages III-IV squamous carcinoma of the vulva. Eight patients were stage III and four were stage IV. Chemotherapy was used as a radiation sensitizer and it was administered in two 5-day cycles 28 days apart. Cisplatin, 50 mg/m2/day iv on Days 1 and 2 or 100 mg/m2 on Day 1 or 2, plus continuous-infusion 5-FU, 1000 mg/m2/day for 4-5 days commencing on Days 1 and 28 of external-beam radiation therapy, are given. The pelvic radiation to a dose of 4400-5400 cGy is administered AP and PA to treat the primary tumor, the groin nodes, and the iliac vessels to the level below the common iliac nodes. Complete tumor responses were seen in 8 of 12 (67%) patients. Responses were observed in 6 of 8 (75%) stage III patients and 2 of 4 (50%) stage IV patients. Partial response were observed in 3 patients, and 1 patient had persistent disease. At the completion of concurrent chemoradiation therapy, radical vulvectomy or excision was used in 3 patients and posterior exenteration in 1. With a median follow-up of 37 months (range, 7-60 months), 10 patients are alive and free of disease, and 2 patients died at 12 and 15 months. There were no treatment-related deaths and no grade 4 toxicity. The morbidity included moist desquamation of the vulva in all patients, with grade 2 toxicity in 10 and grade 3 in 2. One patient had a deep venous thrombosis that responded to anticoagulation therapy. These data support the use of concurrent cisplatin and 5-FU chemotherapy and radiation therapy as an alternative to primary radical surgery to treat advanced-stage squamous carcinoma of the vulva.

Effects of 13 cis-retinoic acid on growth and differentiation of human follicular carcinoma cells (UCLA R0 82 W-1) in vitro.

Dedifferentiation of human thyroid tumors is frequently found in humans. The effect of retinoids (13 cis-RA) was studied on the proliferation and differentiation of a human follicular cell line in vitro (UCLA R0 82 W-1). A significant and dose-dependent reduction (P less than 0.001) in cell number and [3H] thymidine uptake was found in cells exposed to 13 cis-RA up to 10 microM. Higher concentrations of 13 cis-RA, however, led to a dose-dependent restoration of cell proliferation. Various parameters of differentiation increased under the influence of 13 cis-RA (10 microM) over nonexposed cells. The 125I uptake increased 4-fold over that in control nonexposed cells (P less than 0.05). [125I] Epidermal growth factor binding increased 5-fold, and [125I] human TSH binding increased significantly after exposure to 13 cis-RA (P less than 0.02). Deiodinase activity, however, was significantly lower in 13 cis-RA exposed cells than in control cells. The present study shows that 13 cis-RA (10 microM) drives the tumor cells toward a more normal state of proliferation and differentiation.

Olfactory neuroblastoma.

Fifteen patients with olfactory neuroblastoma were treated during the 17-year period of 1969 to 1986. Data was analyzed with respect to age at presentation, sex, presenting signs and symptoms, stage, and results of treatment. Age ranged from 4 to 67 years with the median age being 27 years. Median follow-up was 8 years. Local control was achieved in nine of nine patients or 100% with successful surgical resection, i.e., minimal residual disease, followed by postoperative radiation therapy (45 to 65 Gy) was employed. There were no distant failures when the primary site was controlled. Regional lymph node metastases were infrequent: only 13% (two of 15 patients) presented with positive nodes. Three of four patients treated initially with surgery alone had a local recurrence, two of which were successfully salvaged by combined therapy. There were four patients treated with radiation therapy alone: three had persistent disease after radiation therapy, and one patient was controlled with 65 Gy. Olfactory neuroblastoma has a propensity to recur locally when treated with surgery alone. The authors' experience suggests excellent local control can be achieved with surgery immediately followed by radiation therapy. Thus the authors recommend planned combined treatment for all resectable lesions.

Characterization of a human follicular thyroid carcinoma cell line (UCLA RO 82 W-1).

A thyroid tumor cell line has been established from the metastases of a follicular carcinoma in a female patient. Although the primary tumor released thyroglobulin (Tg) into the circulation (greater than 10,000 ng/ml), the uptake of I131 was less than 2%. After 37 replications the doubling time was 4 days and confluency was reached after 7 days from inoculation of 3 x 10(7) cells. This human thyroid tumor cell line has now been growing in culture for several years. An aneuploid chromosomal pattern was observed (62-82 chromosomes). A pair of X chromosomes was present but no Y chromosome was found which is compatible with the female origin of the cell line. EM studies revealed the presence of microvilli. Immunoperoxidase staining using specific anti-human Tg antisera indicated the presence of Tg within the cells. Nude mice developed solid-cystic tumors within 6 months after injection of the cells. The basal release of immunodetectable Tg, as measured in a perifusion system, increased in response to thyroid stimulating hormone (TSH) (P less than 0.025) or TSH combined with theophylline (P less than 0.001). Unusual isoenzyme patterns for galactose-1-phosphate-uridyltransferase (GALT) and phosphoglucomutase1 (PGM1) were detected in the tumor, compared with normal human fibroblasts and blood cells and isoenzyme patterns from the patient's lymphocytes. Because this malignant human thyroid follicular cell line has retained the ability to synthesize Tg it represents a valuable model for the study of human follicular carcinomas.

Novel antigenic markers of human tumor regression.

The development of tumor-specific antibodies was studied in a group of cancer patients undergoing active specific immunotherapy with irradiated human allogeneic and autochthonous (autologous) tumor cells injected by the intralymphatic route. Immunoblotting studies on extracts of various established tumor cell cultures and fresh tumor biopsies were performed using sera from these patients. Evaluable tumor regressions were associated with detection of antibodies against human tumor cell antigens of 22,000 daltons (22 kd), 38,000 daltons (38 kd), 43,000 daltons (43 kd), and 70,000 daltons (70 kd). Similar antigens of approximately 22, 43, and 70 kd have also been detected in fresh extracts of certain human tumor tissues when tested with antisera from patients responding to immunotherapy. Production of antibodies to these antigens may play a role in tumor regression with active specific immunotherapy. These human regression-associated antigens may, therefore, represent novel agents for cancer immunotherapy.

Early vocal cord carcinoma in obese patients: a surgical lesion.

Early epidermoid carcinoma of the vocal cord is treated successfully by radiation therapy with high 5-year survival rates, low morbidity, and preservation of excellent voice quality in most cases. Typically, surgery is reserved for salvage of radiation failure and provides overall 5-year survival rates of 98% and 90% for T1 and T2 lesions, respectively. The extremely obese patient, often with a short neck and excessive amounts of subcutaneous fat, is difficult for both radiotherapist and surgeon to diagnose and treat. The recent observation of an unusually high rate of radiotherapy failure in a cluster of obese patients with early vocal cord cancer called attention to this problem. Because the larynx is near the thoracic inlet in obese patients, they are not suitable for administration of radiotherapy by accurate opposed lateral portals. The patients reported herein were treated entirely with anterior oblique portals. Of the five obese patients who underwent primary radiation therapy for early vocal cord cancer, three developed recurrent disease (60%) and a fourth developed a severe perichondritis requiring tracheostomy. Two patients with recurrent disease were successfully salvaged with total laryngectomy, while the third patient refused surgery and died 2 years later. The sixth patient was treated by partial laryngectomy with imbrication reconstruction and is alive and without evidence of disease 2 years following surgery. Recurrence rates and complication rates following primary radiation therapy for early vocal cord cancer appear to be unacceptably high in obese patients treated with anterior oblique portals. Therefore, we recommend conservation laryngeal surgery for obese patients with early vocal cord cancer who cannot undergo "standard" radiotherapeutic techniques.

Staging of base of tongue carcinoma by computed tomography.

Staging of carcinoma of the base of the tongue according to the system adopted by the American Joint Committee on Cancer relies on clinical examination possibly augmented by multiple biopsies. Palpation of the tongue base can be difficult without anaesthesia due to retching and vomiting. Computed tomography can, however, accurately depict the deep structures of the base of the tongue without discomfort to the patient. It can also demonstrate the nodal stations of the neck. In 12 patients with primary carcinoma of the base of the tongue the clinical staging results were compared with the CT findings. In 10 of the 12 patients there was good correlation between tumour size and location, while only 2 patients showed a 1.0-1.5 cm discrepancy in the size estimate of the primary tumour. A total of 9 enlarged lymph nodes or nodal groups were only found by CT. The majority of positive nodal stations demonstrated only by CT were in the contralateral neck. Computed tomography is a valuable complement when staging tongue base carcinomas, particularly when evaluating the neck for lymph node metastasis.

Orbital lymphomas and pseudolymphomas: treatment with radiation therapy.

A retrospective analysis of 16 patients with orbital lymphoma or pseudolymphoma from 1961-1984 was undertaken to evaluate the use of radiation therapy. Pathologic assessment confirmed that four patients had benign pseudolymphoma, and 12 patients had true malignant lymphoma, including two with advanced disease at presentation. With a median follow-up of 4 years, the local control rate with radiation therapy was 100%, although the two patients with advanced disease died of lymphoma 26-33 months after irradiation. While doses of 1,600-2,000 cGy appear adequate for pseudolymphoma, for lymphoma a dose of 3,000-4,000 cGy is necessary. Subconjunctival lesions can be treated in a single anteroposterior field; retroorbital lesions require an additional lateral field.

Whole abdominal radiation as salvage therapy for epithelial ovarian cancer.

Thirty patients found to have residual epithelial ovarian cancer at second-look laparotomy were treated with whole abdominal radiation as salvage therapy. Dosage fractions were 120 rad per day until 3000 rad were delivered, then the pelvis was boosted to 5000 rad at 180 rad per day. Fourteen patients (47%) completed therapy without interruption and seven (23%) completed therapy with interruptions due to myelosuppression ranging from one to four weeks. Therapy was not completed in nine patients (30%). Four of 16 patients (25%) with microscopic residual disease before radiation remain alive and free of disease at 22 to 41 months. Two of six (33%) patients with minimal (less than or equal to 5 mm) residual disease remain alive and free of disease 19 to 40 months after radiation treatment. Patients with residual nodules greater than 5 mm uniformly did poorly. Patients who progressed on primary chemotherapy had a median survival of seven months, compared with more than 38 months for chemotherapy responders. Chronic bowel morbidity was a significant problem, with 30% of patients surviving at least four months from completion of radiation requiring laparotomy for small bowel obstruction. These preliminary results suggest that whole abdominal radiation may be useful in the management of patients who have responded to primary chemotherapy, but the benefit is confined to those patients who have minimal or microscopic disease at second-look laparotomy.

Preoperative radiation therapy for locally advanced vulvar cancer.

Eight patients with locally advanced vulvar cancer that would have necessitated pelvic exenteration to encompass the primary tumor were given preoperative radiation therapy in an attempt to shrink the primary tumor and allow more conservative surgery. From 4400 to 5400 rad of external radiation were delivered to the primary tumor, and one patient received an additional 2400 rad from intracavitary therapy. Satisfactory shrinkage of tumor occurred in seven of the eight patients (87.5%), thus allowing conservative surgical excision. In four patients (50%), there was no viable tumor in the surgical specimen. Moist desquamation of the vulva occurred in all patients and was of sufficient severity to require temporary cessation of radiation in four patients (50%). Five received groin radiation, and one (20%) subsequently developed bilateral hip fractures. No other major morbidity occurred. Five of the eight patients (62.5%) are alive without evidence of disease at intervals ranging from 15 months to 10 years. Preoperative radiation in this group obviated the need for pelvic exenteration, resulting in significantly less morbidity without compromising survival.

Induction of canine in vitro reactivity to alloantigen following intralymphatic immunization.

An experimental model has been developed using the dog to study the induction of systemic cell-mediated immunity following intralymphatic immunization (ILI) with allogenic cells. As detected in one-way mixed lymphocyte cultures, blastogenically-reactive immune peripheral blood lymphocytes were observed after the third ILI with 10(7) cells. The in vitro reactivity was augmented by a fourth ILI to a node not previously injected indicating that a response in one node was followed by the trafficing of memory cells to other nodes. No immune PBL were detected after four ILI with lower doses of 10(3) cells. However, these dogs subsequently responded to a single injection of 10(7) cells with high levels of immune lymphocytes which were detectable for up to 24 days. Apparently, ILI with 10(3) or 10(5) cells, while insufficient to produce detectable levels of alloreactive lymphocytes were sufficient for lymphocyte priming. Results obtained with this model will aid in ongoing human trials of intralymphatic immunotherapy of malignant disease.

Ornithine metabolism in normal subjects and patients with cancer.

The metabolism of L-(1-14C)ornithine monohydrochloride was monitored in patients with histologically proven cancer and in normal volunteers. Following i.v. injection of 8 microCi C-14 ornithine (160 nmoles), the decarboxylation of ornithine--yielding 14CO2--was monitored for a 2.5-hr period using the ionization chamber and vibrating-reed electrometer of Tolbert, as modified by Davidson and Schwabe. Twelve normal subjects exhaled 7.3-15.7% of the administered C-14 (mean 12.6% s.d. 3.11%). In ten patients tested before initiation of therapy, recovery ranged from 18.2-32.1% (mean 23.02%, s.d. 4.52%). A t-test indicates a confidence level of > 99.5% that a significant difference exists between the two means. Re-testing of two normal volunteers showed little or no change in ornithine metabolism over a 2-5-mo period. Results from testing three cancer patients before and after therapy correlate well with clinical evidence of the presence of tumor burden.

Distribution and consequences of cell suspensions following intralymphatic infusion.

The fate and consequences of intralymphatic injections of cells was investigated in dogs. The distribution of intact radiolabeled cells was determined in vivo by whole body gamma scanning. Comparison of distributions resulting from intralymphatic, subcutaneous, intradermal and intravenous routes of administration showed that the distribution and duration of radiolabel in various organs varied with the route of administration. Following intralymphatic injection, radiolabel was concentrated in first echelon lymph nodes draining the site of injection and was retained in these nodes for over 4 weeks. Histologic studies showed intense cortical and paracortical lymphopoiesis to be associated with the retention of intralymphatically injected tumor cells by first echelon lymph nodes. Serial histologic examination of lymph nodes from intralymphatically injected inbred beagles revealed that the consequent lymphopoiesis persisted for 5 weeks. In vitro evaluation of peripheral blood and lymph node lymphocyte cytotoxicity to the injected cells indicated that retention and nodal lymphopoiesis was associated with the development of direct lymphocyte cytotoxicity. The effects of concommitant tumor burden, cytotoxic drugs and ionizing radiation were also investigated and suggest that the therapeutic potential for use of the intralymphatic route has not yet been realized.