A phase I study of high-dose ifosfamide and escalating doses of carboplatin with autologous bone marrow support.

The dose-limiting toxicity in two separate phase I trials of the high-dose single agents ifosfamide and carboplatin was renal insufficiency at 18 g/m2 and hepatic and ototoxicity at 2,400 mg/m2, respectively. In this phase I study, 16 adults were treated with ifosfamide at 75% of the single-agent maximum-tolerated dose (MTD) (12 g/m2) and escalating doses of carboplatin (400 to 1,600 mg/m2) to determine the nonhematologic dose-limiting toxicity and the maximum-tolerated dose of the combination. Both drugs as well as mesna for uroprotection were given by continuous infusion over 4 days with an additional day of mesna (total dose per course, 15 g/m2). Autologous bone marrow support was stipulated for subsequent dose levels once granulocytes remained less than 500/microL for more than 14 days in two of three to five patients entered at a given dose level. Autologous bone marrow support was used at doses above the 400 mg/m2 carboplatin dose level. At the maximum-tolerated dose level of 1,600 mg/m2 of carboplatin, renal toxicity precluded further dose escalation. Of the five patients entered at this dose level, reversible creatinine elevation greater than 2 mg/dL (median peak, 2.6 mg/dL) was observed in three patients, and irreversible renal failure occurred in an additional patient (peak creatinine, 6.9 mg/dL. Transient gross hematuria appeared more common with the combination than with ifosfamide alone. Two patients developed severe somnolence and confusion associated with a rising creatinine. There were two complete (CRs) and four partial responses (PRs) in 14 heavily pretreated assessable patients (including four partial or complete responses in eight assessable patients with advanced refractory sarcoma, and one CR in two patients with germ cell carcinoma). Carboplatin and ifosfamide appear to have overlapping renal toxicity. Nevertheless, carboplatin and ifosfamide can be combined at 80% and 75% of the single-agent maximum-tolerated doses, respectively, with acceptable nonhematologic toxicity. Ifosfamide and carboplatin are an attractive core combination for further studies in the treatment of sarcoma, germ cell, ovarian, and lung carcinomas.

A phase I-II study of cyclophosphamide, thiotepa, and carboplatin with autologous bone marrow transplantation in solid tumor patients.

The principles of dose-response and combination chemotherapy were basic to the design of the initial curative standard-dose treatment regimens for leukemias, lymphomas, and testis cancer. Agents were selected with different dose-limiting toxicities, resulting in subadditive toxicity in combination. A fourth principle in the design of curative regimens was to combine agents with different mechanisms of action to avoid cross-resistance. Based on these principles, combinations of the highest tolerated doses of active noncross-resistant agents are required to decrease the emergence of drug resistance and achieve optimum cytotoxicity. Hematopoietic stem-cell support provides a mechanism for significantly increasing the doses of active agents, a strategy that has resulted in the cure of 10% to 50% of selected patients with lymphoma who could not be cured with standard-dose therapy. The lack of sufficiently effective cytoreductive conditioning regimens remains the major impediment to improving the high-dose therapy of patients with solid tumors. In this study, 27 patients with solid tumors were treated with a combination of cyclophosphamide, thiotepa, and carboplatin (CTCb) in a phase I-II study. Severe mucositis and neurotoxicity were dose-limiting. The maximum-tolerated dose (MTD) of the combination was 6.0 g/m2 of cyclophosphamide, 500 mg/m2 of thiotepa, and 800 mg/m2 of carboplatin. There were two deaths (7%) of sepsis, and an overall response rate of 72% in refractory tumors (81% in breast cancer). CTCb is a combination with low morbidity and high cytoreductive efficacy designed to exploit the principles of curative cancer chemotherapy.

High-dose thiotepa alone and in combination regimens with bone marrow support.

Curative treatment regimens for leukemias, lymphomas, and testicular cancer have been based on laboratory observations of a clear relationship (generally linear-log) between increasing doses of chemotherapeutic agents and tumor cytotoxicity and on recognition of the need for combination chemotherapy to avoid the emergence of drug resistance. Chemotherapeutic agents have been selected for combinations based on cytotoxic activity, different mechanisms of action (to avoid cross-resistance), and different dose-limiting toxicities (to avoid additive toxicity). The ideal combinations use the highest tolerable doses of active non-cross-resistant agents to minimize the potential for drug resistance and achieve optimum cytotoxicity. Dose escalation is often limited by myelosuppression. Hematologic stem cell support from bone marrow or peripheral blood allows the administration of significantly higher doses of chemotherapy. In 1977, Thomas and colleagues in Seattle reported that 13 of 100 patients who underwent bone marrow transplantation for relapsed acute leukemia were disease-free 1 to 4.5 years later. Today, almost 50% of selected patients with acute myelogenous leukemia who undergo transplantation with human leukocyte antigen-matched sibling donor marrow during first remission are cured. Between 20% and 50% of lymphoma patients who undergo transplantation after failing conventional treatment have survived; those whose disease is responding to standard-dose therapy at the time of transplant have the best prognosis. Conditioning regimens that are sufficiently cytoreductive are not currently available for patients with solid tumors. The diversity of solid tumors makes it likely that a variety of regimens will be required. In a sequence of laboratory and clinical studies, we have constructed and evaluated a regimen comprising 6 g/m2 of cyclophosphamide, 500 mg/m2 of N,N',N''-triethylenethiophosphoramide (thiotepa), and 800 mg/m2 of carboplatin. The response rate in women with measurable breast cancer was 81%. While profound myelosuppression was noted, organ toxicity has been rare. This regimen, designed to exploit the principles of curative cancer chemotherapy, is associated with low morbidity and high cytoreductive efficacy. The regimen is currently being evaluated in a phase II trial in patients with previously untreated metastatic breast cancer who are responsive to conventional-dose chemotherapy. Of 29 patients entered in the study, only one has died of toxicity, confirming the low incidence of treatment-related toxicity associated with the regimen.

Endogenous digoxin-like immunoreactive factor and digitalis-like factor associated with the hypertension of patients receiving multiple alkylating agents as part of autologous bone marrow transplantation.

1. Hypertension is a complication of autologous bone marrow transplantation when therapy includes multiple alkylating agents. We have sought to identify the factors underlying this hypertension. We measured weight, serum creatinine, plasma renin activity, aldosterone and digoxin-like immunoreactive factor (DLIF), by digoxin radioimmunoassay, in 18 patients. Plasma catecholamines were also measured in five patients. 2. Of the 18 patients studied, 15 became hypertensive. The variable most consistently associated with these individuals' hypertension was DLIF activity which was increased in 14 of the 15 hypertensive patients (P = 0.055, Fisher exact test). Serum creatinine was increased at some point in seven of the 15 hypertensive patients, weight was increased in five and plasma renin activity and aldosterone were increased in one. Catecholamines were not increased in any of the five patients in which they were measured. 3. The association between changes in mean arterial pressure (MAP) and changes in DLIF for the group as a whole was assessed by analysing one data pair per patient, representing the maximal MAP. This correlation was significant (r = 0.75, P = 0.001). 4. Within individual patients, changes in MAP and changes in serum DLIF concentrations were significantly correlated (r greater than 0.50, P less than 0.05) in six of 15 hypertensive patients. 5. Digitalis-like factor (DLF) was measured by inhibition of (Na+,K+)-adenosine 5'-triphosphatase in five patients and DLF and DLIF were significantly correlated (r = 0.081, P = 0.0001). DLF and MAP were also significantly correlated (r = 0.59, P = 0.002). 6. This represents the first longitudinal study of the relationship between DLIF and blood pressure in hypertensive individuals, and the results suggest that DLIF may contribute to the increased blood pressure in some of these subjects.

Cyclophosphamide and thiotepa with autologous bone marrow transplantation in patients with solid tumors.

Autologous bone marrow transplantation to avoid dose-limiting myelosuppression may allow significant drug dose escalations and exploitation of the linear-log correlation between chemotherapy and tumor cytotoxicity. In a phase I trial of cyclophosphamide and thiotepa (with subsequent addition of melphalan), 23 patients were entered; there were two deaths due to toxic effects. The maximum tolerated dose was 6 g of cyclophosphamide/m2 and 720 mg of thiotepa/m2. No significant dose of melphalan could be added. Stomatitis was dose limiting. Eleven of 20 patients who were able to be evaluated responded. Plasma thiotepa concentrations correlated more closely with toxicity and response than with drug dose level. Continuous infusion cyclophosphamide and thiotepa is an active core regimen for the further design of high-dose combination chemotherapy regimens.

Malignant melanoma. Treatment with high-dose combination alkylating agent chemotherapy and autologous bone marrow support.

Nineteen patients with metastatic malignant melanoma were treated with 20 courses of high-dose combination alkylating agent chemotherapy and autologous bone marrow support. All 20 treatment courses were evaluable for toxic reactions and 17 of 20 courses were assessable for response. Twelve of the 20 courses were given at the phase 2 dose per square meter of cyclophosphamide (5.625 g), cisplatin (165 mg), and carmustine (600 mg). Marrow reconstitution occurred with a median time to recovery of 21 and 24 days for more than 500 neutrophils and more than 20,000 platelets, respectively. The overall response rate was 65%, with one patient achieving a complete response with chemotherapy alone. Ten additional patients achieved partial responses following chemotherapy, of which three were subsequently rendered disease free by surgical resection of single areas of residual tumor. Two of these patients are alive and disease free more than 22 months following chemotherapy and one remains relapse free. The median survival for responding patients was 15.2 months and 8.6 months for the entire group.

High-dose combination alkylating agent preparative regimen with autologous bone marrow support: the Dana-Farber Cancer Institute/Beth Israel Hospital experience.

Fifty-nine patients received 61 courses of cyclophosphamide, cisplatin, and carmustine combined. The phase I study consisted of seven dose escalations. Dose levels 5 and 6 also included the attempted addition of melphalan at 40 and 80 mg/m2. The maximum tolerated dose for the three-drug combination was 5620 mg/m2 of cyclophosphamide, 165 mg/m2 of cisplatin, and 600 mg/m2 of carmustine. The dose-limiting toxic effect was fatal veno-occlusive disease of the liver in two of five patients treated at the highest dose level. Veno-occlusive disease of the liver was fatal in two of 40 patients treated at Dose level 4, the maximum tolerated dose. The median time to recovery of polymorphonuclear leukocyte counts to greater than 500/microliters and platelet counts to transfusion independence greater than 20,000/microliters was 19 and 22 days, respectively, after marrow reinfusion. Other toxic effects observed included postdischarge pulmonary toxicity, which appeared to respond to prednisone therapy. Thus, this combination of alkylating agents can be combined at close to the transplant dose of each individual agent. The response rate was high despite considerable prior treatment in most patients. Of 16 evaluable patients with breast cancer, 15 responded (six complete responses). Of six evaluable patients with sarcoma, six responded (one complete response). While patients with melanoma had had no prior treatment, 11 of 17 patients responded (65%). There are currently three patients who are disease-free. Two patients with melanoma were rendered disease-free by excisional biopsy of the only remaining nodule after good partial response, and a patient with metastatic breast cancer remains disease-free after a complete response at 36+ months.

High-dose combination alkylating agent chemotherapy with autologous bone marrow support for metastatic breast cancer.

Seventeen patients with metastatic breast cancer were treated with a high-dose combination chemotherapy regimen and autologous bone marrow support. Thirteen patients had prior combination chemotherapy. Fifteen patients were treated with a phase II regimen of cyclophosphamide (5.625 g/m2), cisplatin (165 mg/m2), and BCNU (600 mg/m2). Bone marrow harvest and reconstitution were uncomplicated. All patients became profoundly myelosuppressed. Fourteen of 16 evaluable patients (88%) responded, including six complete responses (CRs) (38%). The median time to tumor progression was 5 months. The median survival was 8 months. CRs occurred more frequently in patients with no prior chemotherapy for metastatic disease, inflammatory breast cancer; and patients treated within 3 months of first recurrence. The rate of tumor regression was rapid, with a median of 11 days to partial response (PR) and 12 days to CR. Those patients achieving a PR by day 7 had a greater likelihood (P = .03) of attaining a CR than those patients whose PR occurred later. Three deaths (18%) occurred, all in women with inflammatory breast cancer treated with prior chemotherapy. High-dose combined alkylating agent therapy produced high PR and CR rates in metastatic breast cancer patients, most of whom had failed prior chemotherapy. The rate of tumor regression was rapid. Current efforts are directed at developing a regimen using drugs specifically active in breast cancer, with an intent of combining an effective high-dose regimen with additional modalities of therapy in the treatment of breast cancer.

Prediction of the optimal timing of bone marrow reinfusion after high dose chemotherapy.

Autologous bone marrow transplantation allows the use of high dose chemotherapy by obviating dose limiting myelosuppression. The pharmacology of high dose chemotherapy has been inadequately explored, yet this information is critical to determine the timing of marrow infusion and assure that engraftment is not compromised. We have used the Salmonella mutagenesis test (SMT) and colony forming unit-granulocyte, erythrocyte, monocyte, megakaryocyte assay to evaluate the optimal time for marrow infusion after therapy with high dose combinations of alkylating agents (Solid Tumor Autologous Marrow Support Program) in seven patients. The SMT is sensitive, rapidly performed, and has been used to detect mutagenic activity in urine following administration of cyclophosphamide, cisplatin, and 1,3-bis(2-chloroethyl)-1-nitrosourea. In parallel, determination of colony forming ability of the patients own bone marrow (colony forming unit-granulocyte, erythrocyte, monocyte, megakaryocyte assay), when cocultured with autologous serum obtained before and after treatment, provided an assay for circulating marrow toxic drugs or metabolites. The onset of mutagenic activity in the SMT and the in vitro appearance of myelotoxicity by autologous serum in the colony forming unit-granulocyte, erythrocyte, monocyte, megakaryocyte assay were concurrent, and these activities returned to base line at the time of marrow infusion (72 h posttreatment). One patient of the seven was excreting mutagens (TA100 strain only) at the time of marrow reinfusion; he developed hepatic venoocclusive disease, and delayed engraftment. These observations suggest that as high dose regimens evolve the SMT may serve as a rapid, sensitive indicator of the circulation and excretion of toxic compounds, and thereby assist in predicting the optimum time of bone marrow reinfusion.

High-dose combination alkylating agents with autologous bone marrow support: a Phase 1 trial.

Twenty-nine patients were treated with 31 courses of high-dose combination cyclophosphamide, cisplatin, and carmustine (BCNU) with and without melphalan with autologous bone marrow support. Toxicity was dose related. The maximum tolerated dose for cyclophosphamide, cisplatin, and BCNU in this combination in mg/m2 was 5,625, 165, and 600, respectively. Further dose escalation was precluded by the development of multiple organ toxicity, including venoocclusive disease, refractory thrombocytopenia, and hypertension. Melphalan added to the three-drug combination produced excessive renal and gastrointestinal toxicity. Objective tumor regression occurred in 21 of 25 evaluable cases. The results suggest that selected alkylating agents can be combined in full or nearly full doses before nonmyelosuppressive dose-limiting toxicity precludes further escalation.