Immune modulation in autologous bone marrow transplantation: cyclosporine and gamma-interferon trial

From March 1994 to November 1994, 16 patients with high risk hematological malignancies were entered in a phase I clinical trial, designed to confirm the toxicity of cyclosporine and gamma interferon given to induce autologous graft-versus-host disease (GVHD) after autologous bone marrow transplantation (ABMT). This trial was based on the results in a rodent model, in which cyclosporine given after ABMT induces an autoimmune syndrome (autologous GVHD) identical to allogeneic GVHD. Further, this autologous GVHD is associated with a graft-versus-tumor effect augmented by interferon that upregulates MHC class II expression on normal and tumor cells, the target of the cytolytic T cells in autologous GVHD. In this trial, cyclosporine 1 mg/kg/day was given from the day of bone marrow reinfusion until the completion of the interferon and gamma-interferon. Gamma-interferon at 0.025 mg/m2 every other day was started when the total white cell count was >200 cells/ml for 2 consecutive days and continued for a total of 10 doses after ABMT. The preparative regimens were busulfan and cyclophosphamide, or cyclophosphamide with total body irradiation. All patients received 4HC-purged marrow grafts. Median age was 45 years (range 19-68). The diagnoses included chemo-resistant non-Hodgkin's lymphoma (10), acute lymphoblastic leukemia (two), chemo-resistant Hodgkin's disease (two), acute myeloid leukemia (one), and multiple myeloma (one). Median absolute neutrophil count recovery was 25.5 days (range 19-46 days). Median platelet count recovery was 40.5 days (range 28-279 days). There were nine deaths, two were related to transplant toxicity (infection), while the other seven were due to relapse. Event-free survival with a median of 964 days (range 19-1441 days of follow-up was 44%. In conclusion, treatment with cyclosporine, and gamma-interferon after ABMT was well tolerated and did not impair engraftment. Further studies with a larger number of patients are required to document any beneficial anti-tumor effect of autologous GVHD induction after ABMT.

Prospective, randomized, controlled clinical trial comparing traditional versus reduced tidal volume ventilation in acute respiratory distress syndrome patients

OBJECTIVE

To assess the safety and potential efficacy of a mechanical ventilation strategy designed to reduce stretch-induced lung injury in acute respiratory distress syndrome.

DESIGN

Prospective, randomized, controlled clinical trial.

SETTING

Eight intensive care units in four teaching hospitals.

PATIENTS

Fifty-two patients with acute respiratory distress syndrome.

INTERVENTIONS

Traditional tidal volume patients: tidal volume 10-12 mL/kg ideal body weight, reduced if inspiratory plateau pressure was > 55 cm H2O (7.3 kPa). Small tidal volume patients: tidal volume 5-8 mL/kg ideal body weight, to keep plateau pressure < 30 cm H2O (4.0 kPa).

MEASUREMENTS AND MAIN RESULTS

Mean tidal volumes during the first 5 days in traditional and small tidal volume patients were 10.2 and 7.3 mL/kg, respectively (p < .001), with mean plateau pressure = 30.6 and 24.9 cm H2O (3.3 kPa), respectively (p < .001). There were no significant differences in requirements for positive end-expiratory pressure or FIO2, fluid intakes/outputs, requirements for vasopressors, sedatives, or neuromuscular blocking agents, percentage of patients that achieved unassisted breathing, ventilator days, or mortality.

CONCLUSIONS

The reduced tidal volume strategy used in this study was safe. Failure to observe beneficial effects of small tidal volume ventilation treatment in important clinical outcome variables may have occurred because a) the sample size was too small to discern small treatment effects; b) the differences in tidal volumes and plateau pressures were modest; or c) reduced tidal volume ventilation is not beneficial.

Phase I trial to determine the safety, pharmacodynamics, and pharmacokinetics of RSR13, a novel radioenhancer, in newly diagnosed glioblastoma multiforme

PURPOSE

To determine the safety, pharmacokinetics, and pharmacodynamic effect of 2-[4-(3, 5-dimethylanilino)carbonyl]methyl]phenoxy]-2-methylproprionic++ + acid (RSR13) 100 mg/kg/d with radiation therapy (RT) for glioblastoma multiforme (GBM). RSR13, a synthetic allosteric modifier of hemoglobin (HgB), is a novel radioenhancing agent that noncovalently binds to HgB, thereby reducing oxygen binding affinity and increasing tissue oxygen release to hypoxic tissues.

PATIENTS AND METHODS

In this multi-institutional, dose frequency-seeking trial, 19 adult patients with newly diagnosed GBM received RSR13 100 mg/kg every other day or daily along with cranial RT (60 Gy/30 fractions). RSR13 was given over 1 hour by central venous access with 4 L/min of O(2 )by nasal cannula, followed by RT within 30 minutes. Pharmacokinetic (PK) and pharmacodynamic (PD) determinations were performed. The PD end point was shift in P50, the oxygen half-saturation pressure of HgB.

RESULTS

Grade 3 dose-limiting toxicity occurred in none of the patients with every-other-day dosing and in two of the 10 patients with daily dosing. Grade 2 or greater toxicity occurred in three out of nine and six out of 10, respectively. PK and PD data demonstrate that a substantial PD effect was reliably achieved, that PD effect was related to RBC RSR13 concentration, and that there was no significant drug accumulation even with daily dosing. The mean shift in P50 was 9.24 +/- 2.6 mmHg (a 34% increase from baseline), which indicates a substantial increase in tendency toward oxygen unloading.

CONCLUSION

Daily RSR13 (100 mg/kg) during cranial RT is well tolerated and achieves the desired PD end point. A phase II trial of daily RSR13 for newly diagnosed malignant glioma is currently accruing patients within the New Approaches to Brain Tumor Therapy Central Nervous System Consortium to determine survival outcome.

Augmentation of phenylbutyrate-induced differentiation of myeloid leukemia cells using all-trans retinoic acid

Despite preliminary evidence of clinical activity of the putative differentiating agent sodium phenylbutyrate (PB) in the treatment of myeloid neoplasms, it has proven difficult to maintain therapeutic levels of PB above 0.5 mM, well below the ED50 of 1-2 mM. We have studied the impact of combining PB with all-trans retinoic acid (ATRA) on the ML-1 myeloid leukemia cell line. ATRA augmented PB-induced differentiation, cell-cycle arrest, and apoptosis. ATRA augmented PB induction of the myelomonocytic marker CD11b at all doses of ATRA tested (0.0025-1 microM). Although ATRA did not significantly affect the ED50 of PB, the combination of ATRA (1 microM) and PB (0.5 mM) augmented PB-induced CD11b expression eight-fold. Compared to PB alone, this combination of ATRA and PB induced greater cell cycle arrest (S-phase 14% vs 38%; G0/G1-phase cells 72% vs 52%) and greater apoptosis (24% vs 16% by TUNEL assay). Treatment with ATRA (0.5 microM) in combination with PB (0.5 mM) led to significantly greater inhibition of colony formation (4.8% vs 48% inhibition). ATRA combined synergistically with PB to augment CD11b expression and inhibit colony formation. This combination also showed significant interaction in terms of S-phase inhibition. However, this interaction varied as a function of ATRA concentration: antagonistic at low concentrations of ATRA, synergistic at higher concentrations of ATRA. These data suggest that retinoids may significantly augment the cytostatic and differentiating activity of PB, leading to increased potency of the latter drug at clinically achievable doses.

Efficacy of two sputum collection techniques in patients with air flow obstruction

OBJECTIVE

To compare the efficacy of two sputum collection techniques in patients with chronic obstructive pulmonary disease (COPD) in order to diagnose dysplasia or neoplasia.

STUDY DESIGN

This was a crossover study design comparing induced sputum with sputum collected at home. One hundred seven patients with COPD were enrolled. Fifty-six were randomized to collect induced sputum first followed by sputum collection at home. Fifty-one randomly assigned patients collected the sputum in reverse order.

RESULTS

The second sputum collection technique for both random assignments gave the greatest yield of adequate sputum. There was no significant difference in efficacy between the collection of the two sputum collection techniques in the presence of the learning (period) effect.

CONCLUSION

Sputum collection is equally efficacious by the induced method and the home collection method. A learning effect was responsible for the increased yield of sputum abnormalities in the second collection session. Sputum collection at home may facilitate the amount of dysplasic and neoplastic bronchial epithelial changes in heavy smokers with COPD.

Central nervous system cancers in first-degree relatives and spouses

The increasing incidence of high-grade astrocytomas in the elderly, the associations between these malignancies and environmental factors, and case reports suggesting a familial component to these tumors prompted this study of primary brain tumors in first-degree relatives and spouses. This article describes the findings in 154 patients from 72 consecutive families accrued to the National Familial Brain Tumor Registry from 1991 to 1996. Medical records, pathological slides, and demographic data were reviewed for each identified case. Parents and children were affected in 33 families, siblings in 27, and husbands and wives in 12. The median age of the patients was 50.5 years, 55% were men, and 70% had high-grade astrocytomas. The pattern of tumor occurrence in this population is different from most familial cancers. These tumors did not involve multiple generations or occur at an unusually early age. In addition, the cases tended to cluster in time, with 47% of the familial and 50% of the husband-wife cases occurring within a 5-year span. In families with an affected parent and child, the diagnosis was made in the child before the parent in 45% of the cases. Prognostic factors for these patients appear to be similar to that reported for typical high-grade astrocytomas. This study demonstrates that primary brain tumors can occur in families without a known predisposing hereditary disease. The ages of these patients, the clustering of cases in time, the few affected generations, and the occurrence of brain tumors in spouses suggest that environmental exposures may be important in the etiology of this neoplasm. Although this hypothesis requires further study, it is plausible given the known associations in animals and humans between high-grade astrocytomas and radiation, toxic chemicals, and viruses.

The effects of sequential versus concurrent chemotherapy and radiotherapy on survival and toxicity in patients with newly diagnosed high-grade astrocytoma

PURPOSE

To determine the effects of sequential versus concurrent administration of cranial radiotherapy and cisplatin/carmustine (BCNU) chemotherapy on survival and toxicity in newly diagnosed high-grade astrocytomas.

METHODS AND MATERIALS

From 1988 to 1996, 101 patients were treated on 2 therapeutic protocols for malignant glioma that used the identical chemotherapy regimen but differed in the timing of cranial radiotherapy. The eligibility criteria for the 2 protocols were identical. In the first protocol (1988-1991, 52 patients), cisplatin 120 mg/BCNU 120 mg i.v. over 72 h, was given for 3 monthly cycles prior to cranial radiotherapy. After a response rate of 42%, with a median survival of 13 months was achieved with this sequential regimen, a successor protocol (1992-1996, 49 patients) was developed in which cranial radiotherapy began concurrently with the start of the identical chemotherapy regimen. Chemotherapy was delayed but not discontinued if prolonged grade III/IV hematologic toxicity was experienced, but protocol therapy was discontinued if disease progression or thromboembolic events occurred. Survival outcome and hematologic toxicity were compared for the patients treated on these protocols.

RESULTS

Seventy-seven percent of sequentially-treated patients and 68% of concurrently-treated patients completed all planned therapy. Kaplan-Meier survival was similar to concurrent or sequential administration of chemotherapy and radiotherapy (median 12.8 months vs. 13.8 months, respectively). Hematologic toxicity was significantly less in sequentially- versus concurrently-treated patients, with median nadir per cycle (2.9 vs. 1.8 x 10(3)/mm3) (p < 0.001), and incidence of grade 3/4 leukopenia 40% versus 77% (p = 0.002). There was also an increase in platelet transfusion requirements in concurrently-treated patients, but no significant worsening of anemia. We postulate that the worsened leukopenia results from the effects of concurrent radiotherapy on circulating stem cells.

CONCLUSION

Concurrent radiotherapy with this regimen of cisplatin and BCNU chemotherapy did not improve survival, but did increase hematologic toxicity. Therefore, we do not recommend further testing of the concurrent regimen, whereas the sequential regimen is currently under evaluation in a Phase III trial of the Eastern Cooperative Oncology Group and the Southwest Oncology Group. In addition, these studies demonstrate that relatively small radiotherapy fields can deliver a dose to circulating stem cells sufficient to worsen the hematologic toxicity of concurrent myelosuppressive chemotherapy, a phenomena which should be considered in the design of combined modality protocols for other body sites.

Nuclear morphometry adds significant prognostic information to stage and grade for renal cell carcinoma

OBJECTIVES

Identification of patients with a high probability of recurrence after nephrectomy for renal cell carcinoma (RCC) is required for adjuvant studies of new therapies. Nuclear morphometry predicts prognosis for prostate, bladder, and Wilms' tumors and in RCC according to previous small pilot studies.

METHODS

To validate this finding, we studied an additional 101 patients who underwent nephrectomy for Stage pT1 to pT3 RCC at our institution from 1977 to 1993 for whom data regarding recurrence or disease-free survival of greater than 60 months were available. Patient records and pathology specimens were reviewed. Of the 101 patients, 66 (65%) did not experience recurrence with greater than 60 months of follow-up, and 35 (35%) had RCC recurrence with a median time to recurrence of 17 months. Nuclear shape descriptors were tested as predictors of disease recurrence after accounting for stage and grade in proportional hazards regression models.

RESULTS

Range of ellipticity (hazards ratio 3.39, P = 0.014) was confirmed to be a significant predictor of recurrence. A prognostic model using stage, grade, and range of ellipticity identified three distinct groups: low, moderate, and high recurrence risk groups, with recurrence rates of 4%, 37%, and 63%, respectively, at 5 years of follow-up. Morphometry significantly (P = 0.018) improved prognostication on the basis of stage and grade alone in this multivariate model.

CONCLUSIONS

Nuclear morphometry is valid and accurate in predicting relapse in early-stage RCC. The model can select patients with RCC for adjuvant therapies.

CD34+ stem cell augmentation of elutriated allogeneic bone marrow grafts: results of a phase II clinical trial of engraftment and graft-versus-host disease prophylaxis in high-risk hematologic malignancies

Although T cell depletion of allografts used in BMT has reduced GVHD, it has been associated with inferior engraftment and an increased risk of relapse. We have found that T cell depletion by counterflow centrifugal elutriation (CCE) also results in depletion of CD34+ stem cells. In order to determine if the discarded CD34+ cells would improve engraftment, we undertook a phase II trial of allogeneic BMT in which 110 patients (median age 43) with a variety of hematologic malignancies received CD34+ stem cell augmented, elutriated marrow grafts. The T cell-depleted grafts were tightly controlled and contained a mean of 4.3 x 10(7) mononuclear cells/kg, 3.3 x 10(6) CD34+ cells/kg, 1.5 x 10(5) CFU-GM/kg and 5.5 x 10(5) CD3+ T cells/kg. Median time to engraftment of granulocytes (>500/microl) was 16 days and of platelets (>50000/microl) was 25 days, comparable to that seen with unmanipulated marrow. No mixed hematopoietic chimerism was observed that was not associated with disease relapse. The four patients (3.6%) who failed to engraft were all at high risk because of prior donor transfusions or underlying marrow disorders. The incidence of GVHD was dependent on the duration of cyclosporin A (CsA) immunosuppression. In patients who received CsA for > or = 80 days, the incidence of clinically significant acute GVHD (>stage 1) and extensive, chronic GVHD was 5% and 11%, respectively. Peritransplant (< or = 100 day post-BMT) mortality for this group of patients was 15%. Event-free survival in selected subsets of patients compared favorably to previous studies in which patients received unmanipulated marrow allografts.

Increased 9-aminocamptothecin dose requirements in patients on anticonvulsants. NABTT CNS Consortium. The New Approaches to Brain Tumor Therapy

BACKGROUND

High grade astrocytomas remain uniformly fatal despite aggressive surgery and radiotherapy. As existing chemotherapeutic agents are of limited benefit, clinical trials are underway to screen new drugs, such as 9-aminocamptothecin (9-AC), for activity in high grade astrocytomas.

PURPOSE

This study was designed to estimate the efficacy of 9-AC in patients with newly diagnosed glioblastoma multiforme and recurrent high grade astrocytomas. The planned dose of 9-AC for this trial was 850 microg/m2 per 24 h as a 72-h continuous intravenous infusion every 2 weeks. This was the maximum tolerated dose (MTD) on this schedule in multiple phase I studies in patients with systemic malignancies. However, we found this dose subtherapeutic in our patient population. As a result, the purpose of the study was altered to determine the MTD.

METHODS

A group of 32 patients were studied using 850 microg/m2 per 24 h with a provision to escalate to 1000 microg/m2 per 24 h if the first three cycles of 9-AC were without significant hematologic toxicity. Once it was determined that myelosuppression did not occur in patients on anticonvulsants, dose escalations were initiated using the continual reassessment method. Dose escalations were conducted independently in newly diagnosed and recurrent patients and in those taking and not taking hepatic enzyme-inducing anticonvulsants. Pharmacologic studies were conducted during the first cycle of 9-AC. Toxicity was determined using the NCI common toxicity criteria and efficacy was assessed using serial volumetric brain scans.

RESULTS

9-AC was administered to 59 patients, 31 with newly diagnosed glioblastoma multiforme and 28 with recurrent high grade astrocytomas. No grade III-IV myelosuppression was noted in the 29 patients (128 cycles) on phenytoin, carbamazepine, phenobarbital, and/ or valproic acid who received 850 microg/m2 per 24 h. In contrast, two of three patients (five cycles) who were not taking anticonvulsants developed grade IV myelosuppression. Steady-state total 9-AC plasma levels were lower in patients on anticonvulsants (median 25.3 nM) than in patients who were not taking anticonvulsants (median 76.5 nM). Dose escalations performed in 27 additional patients determined the MTD in patients taking anticonvulsants to be 1776 microg/m2 per 24 h for patients with newly diagnosed tumors and 1611 microg/m2 per 24 h for patients with recurrent disease.

CONCLUSIONS

We describe a new and unexpected drug interaction between 9-AC and anticonvulsants. This is similar to recent findings with paclitaxel, and suggests that higher than "usual" doses of some chemotherapeutic agents are required in patients on anticonvulsants. Prospectively defined dose escalations and pharmacologic studies are essential for the careful evaluation of new chemotherapeutic agents in patients with brain tumors.

Polymerase chain reaction-based microsatellite polymorphism analysis of follicular and Hürthle cell neoplasms of the thyroid

Follicular and Hürthle cell carcinomas of the thyroid cannot be differentiated from adenomas by either preoperative fine needle aspiration or intraoperative frozen section examination, and yet there exist potentially significant differences in the recommended surgical management. We examined, by PCR-based microsatellite polymorphism analysis, DNA obtained from 83 thyroid neoplasms [22 follicular adenomas, 29 follicular carcinomas, 20 Hürthle cell adenomas (HA), and 12 Hürthle cell carcinomas (HC)] to determine whether a pattern of allelic alteration exists that could help distinguish benign from malignant lesions. Alterations were found in only 7.5% of informative PCR reactions from follicular neoplasms, whereas they were found in 23.3% of reactions from Hürthle cell neoplasms. Although there were no significant differences between follicular adenoma and follicular carcinoma, HC demonstrated a significantly greater percentage of allelic alteration than HA on chromosomal arms 1q (P < 0.001) and 2p (P < 0.05) by Fisher's exact test. The documentation of an alteration on either 1q or 2p was 100% sensitive and 65% specific in the detection of HC (P < 0.0005, by McNemar's test). In conclusion, PCR-based microsatellite polymorphism analysis may be a useful technique in distinguishing HC from HA. Potentially, the application of this technique to aspirated material may allow this distinction preoperatively and thus facilitate more optimal surgical management. Consistent regions of allelic alteration may also indicate the locations of critical genes, such as tumor suppressor genes or oncogenes, that are important in the progression from adenoma to carcinoma. Finally, this study demonstrates that Hürthle cell neoplasms, now considered variants of follicular neoplasms, differ significantly from follicular neoplasms on a molecular level.

Practical implementation of a modified continual reassessment method for dose-finding trials

PURPOSE

We describe a practical, reliable, efficient dose-finding design for cytotoxic drugs applied in a multi-institutional setting.

METHODS

The continual reassessment method (CRM) was modified for use in phase I trials conducted through the New Approaches to Brain Tumor Therapy (NABTT) Consortium. Our implementation of the CRM uses (1) a simple dose-toxicity model to guide data interpolation, (2) groups of three patients to minimize calculations and stabilize estimates, (3) investigators' clinical knowledge or opinion in the form of data to make the process easier to understand, and (4) a flexible computer program and interface to facilitate calculations.

RESULTS

The modified CRM was used in two dose-finding trials of 9-aminocamptothecin in patients with newly diagnosed and recurrent glioblastoma who were taking anticonvulsant medication. The CRM located the maximum tolerated dose (MTD) efficiently in both trials. Compared to conventional designs, the CRM required slightly more than half the number of patients expected, did not greatly overshoot the MTD (i.e. no patients were treated at dangerously high doses), and did not underestimate the MTD.

CONCLUSIONS

Our experience demonstrates the feasibility of implementing this design in multi-institutional trials and the possibility of performing dose-finding studies that require fewer patients than conventional methods.

DNA content and other factors associated with ten-year survival after resection of pancreatic carcinoma

BACKGROUND AND OBJECTIVES

The 5-year survival rates after resection of pancreatic carcinoma have recently increased and are predicted by tumor size, DNA content, and lymph node metastases at the time of resection. However, whether the 10-year survival rates have also increased and are similarly predicted by these factors is not known.

METHODS

The influence of preoperative imaging tests, alcohol consumption, cigarette smoking, K-ras mutations, anatomic location, details of surgical resection, pathologic findings, and tumor DNA content on survival was tested for 96 patients after a successful resection of a pancreatic carcinoma with 17 patients being followed for more than 5 years.

RESULTS

The 5- and 10-year patient survival rates were 18% and 3%, respectively. Univariate and multivariable analyses showed that tumor DNA content, pathologic tumor size, and lymph node metastases were the strongest prognostic indicators for long-term patient survival, although the importance of tumor size may diminish 2 or more years after resection. Surprisingly, the 11 patients with diploid carcinomas > or = 4 cm had an estimated 10-year survival rate of 36%.

CONCLUSION

These results show that the 10-year survival rate for pancreatic carcinoma remains very low, although the subset of patients with biologically favorable tumors has a prolonged survival and possible cure after resection.

A randomized controlled trial comparing the cuffed oropharyngeal airway and the laryngeal mask airway in spontaneously breathing anesthetized adults

BACKGROUND

The cuffed oropharyngeal airway (COPA), a modified Guedel airway, was compared with the laryngeal mask airway (LMA) during spontaneous breathing anesthesia. Specifically examined were ease of use, physiologic tolerance, and the frequency of problems.

METHODS

Adult patients consented to random (2:1) assignment to either COPA (n = 302) or LMA (n = 151) for airway management during anesthesia with propofol, nitrous oxide, and oxygen.

RESULTS

Ease of insertion was similar, but the first-time successful insertion rate was higher with the LMA (COPA, 81% compared with LMA, 89%; P = 0.05). More brief manipulations (head tilt, chin lift, jaw thrust) were reported in the COPA group (average total number of manipulations: COPA, 1.1 +/- 1.6 compared with LMA, 0.1 +/- 0.2; P < 0.001). Continuous airway support was used more frequently in the COPA group (COPA, 30% compared with LMA, 0%; P < 0.0005). The incidences of aspiration, regurgitation, laryngospasm, wheezing, succinylcholine administration, oxygen saturation (SpO2) < 92%, failed use, and minor intraoperative problems were similar. When the airways were removed, blood was detected on the COPA less frequently than on the LMA (COPA, 5.8% compared with LMA, 15.3%; P = 0.001). The incidence of early and late sore throat was greater with the LMA (early: COPA, 4.7% compared with LMA, 21.9% [P = 0.001]; late: COPA, 8.4% compared with LMA, 16.1%; P = 0.01). The LMA did better than the COPA when anesthetists analyzed the technical aspects of the two devices.

CONCLUSIONS

Although the COPA and LMA are equivalent devices in terms of physiologic alterations and overall clinical problems associated with their use, the LMA was associated with a higher first-time insertion rate and fewer manipulations, suggesting that it is easier to use. The COPA was associated with less blood on the device and fewer sore throats, suggesting it may cause less pharyngeal trauma. Ultimately, both devices were similar in establishing a safe and effective airway for spontaneously breathing anesthetized adults.

DNA content and other factors associated with ten-year survival after resection of pancreatic carcinoma

BACKGROUND AND OBJECTIVES

The 5-year survival rates after resection of pancreatic carcinoma have recently increased and are predicted by tumor size, DNA content, and lymph node metastases at the time of resection. However, whether the 10-year survival rates have also increased and are similarly predicted by these factors is not known.

METHODS

The influence of preoperative imaging tests, alcohol consumption, cigarette smoking, K-ras mutations, anatomic location, details of surgical resection, pathologic findings, and tumor DNA content on survival was tested for 96 patients after a successful resection of a pancreatic carcinoma with 17 patients being followed for more than 5 years.

RESULTS

The 5- and 10-year patient survival rates were 18% and 3%, respectively. Univariate and multivariable analyses showed that tumor DNA content, pathologic tumor size, and lymph node metastases were the strongest prognostic indicators for long-term patient survival, although the importance of tumor size may diminish 2 or more years after resection. Surprisingly, the 11 patients with diploid carcinomas > or = 4 cm had an estimated 10-year survival rate of 36%.

CONCLUSION

These results show that the 10-year survival rate for pancreatic carcinoma remains very low, although the subset of patients with biologically favorable tumors has a prolonged survival and possible cure after resection.

The New Approaches to Brain Tumor Therapy (NABTT) CNS Consortium: Organization, Objectives, and Activities

BACKGROUND: Despite advances in neuro-imaging, neurosurgery, radiation therapy, and chemotherapy, limited progress has been made in the treatment of patients with high-grade astrocytomas. The National Cancer Institute has attempted to speed advances in this field by funding CNS consortia to conduct innovative clinical trials in this patient population since 1994. METHODS: The NABTT CNS Consortium is composed of a consortium headquarters and nine member institutions with outstanding multidisciplinary expertise, clinical and laboratory research capabilities, and access to large numbers of patients with brain tumors. RESULTS: The objectives of the NABTT Consortium are to improve the therapeutic outcome for adults with primary brain tumors, to conduct basic science and clinical research, and to improve the care and quality of life of adults with primary brain tumors. NABTT's clinical studies have discovered important drug interactions between anticonvulsant and antineoplastic agents, defined the activity of paclitaxel and 9-aminocamptothecin in glioblastoma multiforme, tested a novel dose escalation strategy for brain tumor trials, and established new protocol "classes" to expedite and standardize clinical research in this field. CONCLUSIONS: Significant progress in the care of patients with primary brain tumors is likely to result from the highly focused and multidisciplinary efforts of the NIH-funded CNS consortia.

Immunomodulatory properties of antineoplastic drugs administered in conjunction with GM-CSF-secreting cancer cell vaccines

Cancer cells genetically modified to secrete immunoregulatory cytokines offer great promise for human cancer treatment as tumor vaccines. However, in preclinical animal studies, large established cancer burdens have appeared difficult to eradicate with such vaccines. For example, lethally-irradiated GM-CSF-secreting CT26 colon carcinoma cell vaccine therapy tends to cure only animals bearing 1 x 10(5) wild-type CT26 cells or less. For many human cancers, antineoplastic chemotherapy can often significantly reduce systemic cancer burdens. Unfortunately, for most advanced metastatic solid organ cancers, such as cancers of the breast, colon, and prostate, antineoplastic drug treatments generally fail to effect cancer cures. Treatment regimens combining genetically-modified cancer cell vaccine therapy and antineoplastic chemotherapy have the potential to increase advanced cancer cure rates if antineoplastic drugs and drug combinations that do not inhibit vaccine-induced immune responses can be identified. To assess the potential immunomodulatory properties of commonly-used antineoplastic drugs that might be used in combination with cancer vaccine treatments, we studied the effects of the drugs on antitumor immune responses manifest by animals receiving lethally-irradiated GM-CSF-secreting CT26 cell vaccines. Immunomodulatory properties of the antineoplastic drugs were evaluated i) by monitoring drug effects on the generation of tumor-specific CD8+ cytotoxic T-lymphocytes (CTLs) in response to GM-CSF-secreting CT26 vaccine administration, ii) by determining drug effects on the resistance of vaccinated animals to subsequent challenge with lethal inoculac of CT26 cells, and iii) by evaluating combination drug and vaccine treatment efficacy against established CT26 tumors. Using this approach, doxorubicin was found to possess apparent immunostimulatory activities, depending on the dose and schedule of administration, while cyclophosphamide appeared immunosuppressive. The different immunomodulatory properties of doxorubicin and cyclophosphamide may be clinically relevant: combination doxorubicin and vaccine treatment of established CT26 cancers increased cure rates over that achieved with either agent alone, while combination cyclophosphamide and vaccine treatment of animals carrying CT26 tumors was no better in curing the animals than drug treatment alone.

Prospective detection of preclinical lung cancer: results from two studies of heterogeneous nuclear ribonucleoprotein A2/B1 overexpression

The United States lung cancer epidemic has not yet been controlled by present prevention and treatment strategies. Overexpression of a Mr 31,000 protein, heterogeneous nuclear ribonucleoprotein (hnRNP) A2/B1, had shown promise as a marker of lung cancer. In a pilot study of archived preneoplastic sputum specimens, hnRNP A2/B1 overexpression more accurately detected preclinical lung cancer than standard cytomorphology. In separate, ongoing prospective studies, sputum is collected annually from stage I resected non-small cell lung cancer patients at high risk of developing a second primary lung cancer and Yunnan tin miners at high risk of primary lung cancer. After the first year of follow-up, preclinical detection of lung cancer by routine cytology was compared with hnRNP A2/B1 overexpression as measured by quantitative densitometry of immunostained slides. Up-regulation of hnRNP A2/B1 in sputum specimens accurately predicted the outcome in 32 of 40 primary lung cancer and control patients within 12 months, whereas cytological change suggestive of lung cancer was found in only 1 patient. In the primary lung cancer study, overexpressed hnRNP A2/B1 accurately predicted the outcome in 69 of 94 primary lung cancer and control miners, whereas only 10 with primary lung cancer were diagnosed cytologically. These two prospective studies accurately predicted that 67 and 69% of those with hnRNP A2/B1 up-regulation in their sputum would develop lung cancer in the first year of follow-up, compared with background lung cancer risks of 2.2 and 0.9% (35- and 76-fold increase, respectively). Using sputum cells to monitor hnRNP A2/B1 expression may greatly improve the accuracy of preclinical lung cancer detection.

Preirradiation paclitaxel in glioblastoma multiforme: efficacy, pharmacology, and drug interactions. New Approaches to Brain Tumor Therapy Central Nervous System Consortium

PURPOSE

The purpose of this study was to determine the response rate of paclitaxel administered at maximal tolerated doses (MTD) in patients with newly diagnosed glioblastoma multiform.

PATIENTS AND METHODS

All patients in this multicenter study were 45 years or older and had measurable residual tumor on postoperative MRI scans. Up to 3 cycles of paclitaxel were administered as a continuous 96-hour intravenous infusion prior to radiation, provided that the tumor did not enlarge on serial MRIs. The initial 10 patients were treated with the previously recommended phase II dose of 140 mg/m2. Less than anticipated toxicity led to the development of a phase I/II study in 24 patients in which paclitaxel doses were escalated separately in patients receiving (+EIAED) or not receiving (-EIAED), concomitant enzyme-inducing antiepileptic drugs. Paclitaxel plasma steady-state concentrations (Css) were measured during the first cycle of chemotherapy. Response was the primary efficacy endpoint for this study, although survival was also assessed.

RESULTS

The MTD was 140 mg/m2 in the -EIAED, and 200 mg/m2 in the +EIAED patient groups. The mean Css for the -EIAED patients treated at 140 mg/m2 was 38 nM, whereas the mean Css for +EIAED patients were 17 nm at 140 mg/m2, 27 nM at 175 mg/m2, 46 nM at 200 mg/m2, and 51 nM at 230 mg/m2. One patient, who had a verified partial response, had his diagnosis changed to an anaplastic oligodendroglioma on subsequent central neuropathologic review. None of the 15 assessable glioblastoma patients treated at or above the MTD doses showed a radiographic response to paclitaxel. The median survival of eligible patients on this protocol was 355 days (95% CI, 255 to 485 days), which is similar to the survival of comparable patients treated with conventional therapy.

CONCLUSIONS

These results suggest that (1) paclitaxel given as a 96-hour infusion at the MTD has minimal activity in patients with untreated glioblastoma, (2) the concomitant administration of EIAEDs alters the pharmacology of paclitaxel, resulting in a lower Css, reduced systemic toxicity, and higher dose requirements, (3) this study design, in which a new agent is given prior to radiation therapy (with serial monitoring of MRI), did not adversely affect survival in this patient population.

Androgen receptor variants with short glutamine or glycine repeats may identify unique subpopulations of men with prostate cancer

The androgen receptor (AR) contains glutamine (CAG) and glycine (GGC) repeats that are each polymorphic in length. We screened clinically localized prostate cancers for somatic mutations in the length of the CAG and GGC repeats in the AR gene and characterized the length of these repeats in the germ-line AR gene. Somatic mutations were rare, and the range of germ-line repeat lengths in men with prostate cancer was within the range of normal in the general population. Most allele frequencies in Caucasian men with clinical prostate cancer were remarkably comparable to those in the general Caucasian population. However, a subpopulation of the men with clinical prostate cancer had a substantially higher frequency of AR alleles with 16 or 17 CAGs (6 of 59 men, 10%) than did the general population (6 of 370 alleles, 1.6%), and a different subpopulation of the men with prostate cancer had a higher frequency of AR alleles with 12 or 13 GGCs (7 of 54 men, 13%) than did the general population (1 of 110 alleles, 0.9%). Of the men with prostate cancer who had an AR gene with 16 or 17 CAGs, 83% had lymph node-positive disease, despite the lack of clinical evidence of metastatic spread. This suggests that a short AR CAG allele may be a risk factor for the development of clinically unsuspected lymph node-positive prostate cancer among men undergoing radical prostatectomy and raises the question of whether this short repeat length played an active role in the development of aggressive prostate cancer. The odds of having a germ-line AR gene with a short CAG repeat (</=17 CAGs) were substantially higher in Caucasian men with lymph node-positive prostate cancer than in Caucasian men with lymph node-negative disease or in the general Caucasian population. The odds of having a short germ-line AR CAG were the same for men with lymph node-negative prostate cancer as for the general Caucasian population. The odds of having a germ-line AR gene with a short glycine repeat (</=14 GGCs) were substantially higher in men with prostate cancer than in the general population, but the frequency of alleles with a short GGC repeat was the same in men with lymph node-positive versus lymph node-negative disease. This suggests that a short GGC repeat may be a risk factor for the development of clinical prostate cancer, a hypothesis that needs to be tested in cohort and case-control studies.

Phase II study of continuous infusion carmustine and cisplatin followed by cranial irradiation in adults with newly diagnosed high-grade astrocytoma

PURPOSE

To evaluate the activity and toxicity of carmustine (BCNU) and cisplatin administered as a 72-hour continuous intravenous infusion before radiation in adults with newly diagnosed high-grade astrocytomas.

PATIENTS AND METHODS

Fifty-two patients with a Karnofsky performance status greater than 60 and no prior antineoplastic therapy entered this protocol. The median age of the patients was 55 years. Eighty-eight percent had glioblastoma multiforme and 12% had anaplastic astrocytomas. BCNU (40 mg/m2/d) and cisplatin (40 mg/m2/d) were administered concurrently as a 72-hour infusion every 3 to 4 weeks. Radiation was begun 4 weeks after the third cycle of chemotherapy or earlier for progressive disease. Responses required a > or = 50% reduction in contrast-enhancing volume.

RESULTS

Forty patients (77%) completed three chemotherapy infusions, five (10%) received two infusions, and seven (13%) received only one. Fifty-one patients completed radiation. Seventeen (42%) patients with measurable disease had a partial response (PR) to chemotherapy, 23 (53%) had stable disease (SD), and two (4%) had progressive disease (PD) on chemotherapy. The median survival time for all patients was 13 months. Survival rates at 1, 2, 3, and 5 years were 62%, 19%, 12%, and 5%, respectively. Grade III to IV leukopenia occurred in 32% of patients; 63% received platelet transfusions and 58% required RBCs. Neutropenic fevers were rare and no intracranial hemorrhages or treatment-related deaths were noted. Nausea, vomiting, peripheral neuropathy, hearing loss, and thromboembolic events were relatively common.

CONCLUSION

This chemotherapy regimen appears to have significant activity and may prolong survival in adults with newly diagnosed high-grade astrocytoma.

A new subarachnoid hemorrhage grading system based on the Glasgow Coma Scale: a comparison with the Hunt and Hess and World Federation of Neurological Surgeons Scales in a clinical series

OBJECTIVE

Although the Hunt and Hess Scale (HHS) and World Federation of Neurological Surgeons Scale (WFNSS) are the most widely used subarachnoid hemorrhage (SAH) grading systems, neither system has achieved universal acceptance. We propose a simplified grading system based entirely on the Glasgow Coma Scale (GCS), which compresses the 15-point GCS into five grades that are comparable with those of the HHS and WFNSS. We refer to this system as the GCS grading system and present a direct comparison with the HHS and WFNSS for predictive value regarding patient outcome and interrater reliability.

METHODS

We reviewed 291 consecutive patients with aneurysms treated at our institution between January 1992 and January 1996 and compared the admission grades from the GCS, WFNSS, and HHS with outcome measures at discharge from hospitalization. The Glasgow Outcome score was used as the major outcome measure to evaluate the predictive value of the three scales. Mortality and length of stay (LOS) were also evaluated as outcome measures. The predictive value of each scale was tested with an ordinal logistic regression model for Glasgow Outcome score, a logistic regression model for mortality data, and a linear regression model for LOS.

RESULTS

Using the logistic regression model, the GCS was the best predictor of discharge Glasgow Outcome score, with an odds ratio of 2.585 (P = 0.0001), compared with 2.311 (P = 0.0001) for the WFNSS and 2.262 (P = 0.0001) for the HHS. Using mortality data in the logistic model, the HHS was the best predictor, with an odds ratio of 3.391 (P = 0.0001), compared with 2.859 (P = 0.0001) for the GCS and 2.560 (P = 0.0001) for the WFNSS. Each of the three scales had a high predictive value for LOS, using a linear model. We discuss, however, the problematic nature of LOS as an outcome measure for SAH. Interrater reliability for each scale was evaluated using kappa statistics, based on 15 additional patients evaluated prospectively, and showed that the GCS grade also had the greatest interrater reliability, with a kappa of 0.46 (P = 0.0002), compared with 0.41 (P = 0.0005) for the HHS and 0.27 (P = 0.027) for the WFNSS.

CONCLUSION

We conclude that the GCS grade has equal or greater predictive value regarding outcome after SAH than do the currently used grading systems and that it has greater reproducibility across observers. Broader familiarity with the GCS among medical and paramedical personnel may further enhance the usefulness of the GCS grade over the HHS and WFNSS in providing a standardized, universally accepted grading system for SAH.