Hydrazine Sulfate Influence on Nutritional Status and Survival in Non-Small-Cell Lung Cancer
[Journal of Clinical Oncology 8:9-15, 1990]
By Rowan T. Chlebowski, Linda Bulcavage, Mary Grosvenor, Engin Oktay, Jerome B. Block, Joan S. Chlebowski, Ishrat Ali, and Robert Elashoff
Department of Medicine, Division of Medical Oncology, Harbor-UCLA Medical Center, Torrence, CA; Johnson Comprehensive Cancer Center, UCLA School of Medicine, Los Angeles, CA; Southern California Permanente Medical Group, Los Angeles and Bellflower, CA.
This randomized, prospective, placebo-controlled clinical trial compares the influence on nutritional status and survival of hydrazine sulfate with placebo addition to cisplatin-containing combination chemotherapy in patients with unresectable non-small-cell lung cancer (NSCLC). The trial consisted of 65 patients with advanced, unresectable NSCLS who had had no prior chemotherapy, were at least partially ambulatory (Eastern Cooperative Oncology Group [ECOG] performance status [PS] level 0-2), and who had adequate hematologic, renal, and hepatic function. All patients received the same defined combination chemotherapy (cisplatin, vinblastine, and bleomycin) and the same defined dietary counseling with the addition of either three times daily oral hydrazine sulfate (60 mg) or placebo capsules. Hydrazine sulfate compared with placebo addition to chemotherapy resulted in significantly greater caloric intake and albumin maintenance (P < .05). considering all patients, survival was greater for the hydrazine sulfate compared with placebo group (median survival, 292 v 187 days), but the difference did not achieve statistical significance. In favorable PS patients (PS 0-1), survival was significantly prolonged (median survival, 328 days v 209 days; P < .05) for hydrazine sulfate compared with placebo addition. In a multifactor analysis, PS, weight loss, and liver involvement were the final variables. Objective response frequency and toxicity were comparable on both arms. Hydrazine sulfate may favorable influence nutritional status and clinical outcome of patients with NSCLS. Further definitive studies of hydrazine sulfate addition to therapeutic regimens in NSCLC are warranted.
J Clin Oncol 8:9-15. © 1990 by American Society of Clinical Oncology.
This randomized, prospective, placebo-controlled clinical trial compares the influence on nutritional status and survival of hydrazine sulfate with placebo addition to cisplatin-containing combination chemotherapy in patients with unresectable non-small-cell lung cancer (NSCLC). The trial consisted of 65 patients with advanced, unresectable NSCLC who had had no prior chemotherapy, were at least partially ambulatory (Eastern Cooperative Oncology Group [ECOG] pe4ormance status [PS] level 0-2), and who had adequate hematologic, renal, and hepatic function. All patients received the same defined combination chemotherapy (cisplatin, vinblastine, and bleomycin) and the same defined dietary counseling with the addition of either three times daily oral hydrazine sulfate (60 mg) or placebo capsules. Hydrazine sulfate compared with placebo addition to chemotherapy resulted in significantly greater caloric intake and albumin maintenance (P < .05). Considering all patients, survival was greater for the hydrazine sulfate compared with placebo group (median survival, 292 vs. 187 days), but the difference did not achieve statistical significance. In favorable PS patients (PS 0-1), survival was significantly prolonged (median survival, 328 days v 209 days; P < .05) for hydrazine sulfate compared with placebo addition. In a multifactor analysis, PS, weight loss, and liver involvement were the final variables. Objective response frequency and toxicity were compatible on both arms. Hydrazine sulfate may favorably influence nutritional status and clinical outcome of patients with NSCLC. Further definitive studies of hydrazine sulfate addition to therapeutic regimens in NSCLC are warranted.
The treatment of patients with unresectable non-small-cell lung cancer (NSCLC) remains a formidable challenge. Radiation therapy alone does not influence overall survival. Chemotherapy, especially with cisplatin combinations, has had extensive evaluation with measurable but modest improvement in outcome. Increases in median survival of between 3 to 21 weeks have been reported in randomized trials comparing cisplatin combination chemotherapy to supportive care alone. In addition, comparative trials have failed to establish a clear survival advantage for any one chemotherapy program.
However, other recent studies have demonstrated: (1) both hypoalbuminemia and altered nutritional status to be associated with poor prognosis in patients with NSCLC; (2) failure of caloric provision alone to improve clinical outcome of this group; (3) abnormal glucose and protein metabolism to be commonly present in cancer patients with altered nutritional status ; and, finally, that (4) hydrazine sulfate, an inhibitor of gluconeogenesis in animals improves these glucose and protein metabolic abnormalities with associated maintenance of albumin and weight in cancer populations. Based on these results, we initiated a randomized, prospective, placebo-controlled clinical trial to evaluate whether addition of hydrazine sulfate compared with placebo results in significant improvement in clinical outcome in terms of nutritional status and survival of patients with advanced NSCLC receiving the same defined cisplatin-containing chemotherapy regimen.
Patients and Methods
Patients were required to have histologically confirmed NSCLC with either extensive disease (American Joint Commission [AJC] stage 111, M) or bulky limited disease (AJC stage Ill, M) that was unsuitable for curative surgery or radiotherapy. In the new International Staging System for lung cancer, the patients would have had stage IV or stage IIIb diseases respectively.
Patients whose tumors were, in whole or part, small-cell anaplastic carcinoma were excluded. Patients had to be at least partially ambulatory (Eastern Cooperative Oncology Group [ECOG] performance status [PS] level 0, 1, or 2), had to have received no prior chemotherapy, and had to have measurable or evaluable disease. Patients with prior radiotherapy were eligible if there was at least one measurable or evaluable site outside radiation fields; if measurable or evaluable disease was present within radiation fields, the disease had to have shown clear-cut progression, and at least 3 months had to have elapsed since completion of radiotherapy. Patients with documented CNS involvement requiring corticostcroid therapy were excluded. Adequate bone marrow, renal, and hepatic function were required; WBC count greater than 3,000/mL; granulocytes greater than 1,200/ mL; platelets greater than 100,000/mL; bilirubin less than 2 mg/dL; and serum creatinine less than 2 mg/dL. These entry criteria are closely comparable to those in use for nearly a decade in Cancer Cooperative Group NSCLC trials and define a relatively homogeneous patient population that most practicing oncologists would consider for therapy.
Additional exclusion criteria based on potential influence on nutrition end points included diabetes mellitus requiring insulin and presence of ascites or clinically significant edema (to avoid confounding weight determinations). All patients gave informed written consent.
Study Design and Treatments
Hydrazine Sulfate was evaluated in this study under investigational new drug (IND) 17,671 from the Food and Drug Administration (FDA) as a randomized, prospective, double-blind, placebo-controlled trial. Patient randomization was conducted using published random number tables. A hospital pharmacist held the code for the trial, thus ensuring that the patient and treating physician remained blinded to the treatment allocation. Patients were randomly assigned to receive the same defined chemotherapy regimen with either placebo or hydrazine sulfate addition according to the outline shown in Table 1.
In both groups, chemotherapy was given for a total of six cycles or until disease progression occurred. Concurrent radiotherapy was not permitted. Hydrazine/placebo use was continued in the absence of limiting toxic effects until disease progression occurred. Chemotherapy dosage was adjusted in a defined fashion on the basis of day-of-treatment blood counts. There were also dose reductions for renal toxicity, severe nausea and vomiting, hepatic dysfunction, and neurotoxicity. All antiemetic use was recorded.
The ECOG toxicity and standard response criteria for complete and partial response were used. Patients were evaluated for response at the start of each cycle of chemotherapy by clinical examination and by chest x-ray. To allow assessment of any potential influence of hydrazine sulfate on cytotoxic chemotherapy delivery, the actual chemotherapy dose administered was compared with the dose mandated by the protocol and recorded as a percentage for the initial two courses. Use of radiotherapy "off study" after disease progression was recorded in all cases to evaluate any possible influence of local chest radiotherapy on long-term survival.
The same defined dietary counseling was provided on both treatment arms to ensure comparability of dietary information available to patients on both hydrazine and placebo treatments. The content of the monthly dietary counseling session was based on patients' nutritional status at entry. The dietary counseling was designed to duplicate the routine clinical dietary consultation that would be a component of a cancer patient's standard clinical management. The goals of the dietary counseling given to all patients in this study were to prevent weight loss (in patients 95% of their pre-illness weight) and promote weight gain (in those patients < 95% of their pre-illness weight). The target caloric intake was determined as 1.2 x basal energy expenditure ([BEE], calculated from height, weight, age, and sex) for weight maintenance and 1.5 x BEE for weight gain. Only dietary counseling techniques were used to approach this target intake; enteral tube feedings or parenteral nutrition support was not given to any patient while on study.
All nutritional assessments and the initial two chemotherapy courses were administered in the Harbor-UCLA Clinical Research Center. Before each chemotherapy course, body weight was measured on the same printing scale. At the same interval, serum albumin was assessed and caloric intake was estimated using a 24-hour dietary recall. Statistically significant differences between patient groups receiving hydrazine and placebo therapy relative to pretreatment clinical factors were sought using chi-sqaured contingency table analysis and Student's t-test. The paired t-test (two-tailed) was used to analyze differences over time in the same patient. Response frequency between arms was compared using the chi-squared test. Data are expressed as mean ± SEM unless otherwise stated. Survival curves were constructed using the Kaplan-Meier method with survival for treatment groups compared using the log-rank test. No interim analyses were performed. Survival time was calculated from the date of randomization until death. All hypothesis testing was two-sided. Statistical comparisons of survival differences include all randomized patients. A Weibull model was used for multivariate analyses.
The trial met its prospectively determined target size of 60 patients and was designed so that a 35% difference in survival at 1 year could be detected with a power of 0.8 at the 0.05 level of significance. Prior metabolic studies indicated that approximately 70% of patients with NSCLC improved their metabolic status following hydrazine sulfate use. The sample size estimate was based on the anticipation that half of the patients who improved their metabolic status would achieve clinical benefit.
A total of 65 patients were randomized with all evaluable for toxicity and response. Pretreatment patient characteristics of placebo and hydrazine groups were comparable including age, sex, PS, prior weight loss, lactic dehydrogenase (LDH), and disease extent (bulky limited v extensive; percentage with bone and/or liver metastases) (Table 2). In the important PS 0-1 subgroup, pretreatment patient characteristics also were not significantly different; the same number of patients had bone (two per arm) or liver (one per arm) metastases in each group. The only patient who had brain metastasis was in the hydrazine arm.
The survival of all patients given hydrazine sulfate or placebo in addition to combination chemotherapy is outlined in Fig 1. Considering all patients, median survival was greater for the hydrazine versus placebo group (292 days vs. 197 days, respectively), but the difference did not achieve statistical significance (P = .11). As seen in Fig 2, survival benefit of hydrazine sulfate occurred in the favorable PS group (PS 0-1). For PS 2 patients (those bedridden at least part of the day), survivals were short (median, 132 days) and closely comparable for hydrazine sulfate or placebo. For the favorable PS patients (PS 0-1), survival was significantly prolonged with hydrazine sulfate compared with placebo addition (median survival, 328 days v 209 days, respectively; P < .05, a 17-week increase). The survival at I year in the PS 0-1 patients also was significantly increased (P<.05) for hydrazine compared with placebo addition (42% alive v 18%, respectively). The significance levels for survival in a multifactor analysis are given in Table 3 for both the entire patient population and the PS level 0- I patients. In the adjusted multifactor model, PS, weight loss, and liver involvement were the final variables.
The influence of hydrazine sulfate on nutritional status is illustrated in Table 4. As can be seen, hydrazine sulfate resulted in significantly greater (P < .05) albumin maintenance compared with placebo treatment. Caloric intake was also increased (P < .05) in the hydrazine group. The moderate increase in body weight associated with hydrazine use did not achieve statistical significance. Since discontinuation of smoking could potentially have an impact on nutritional end points through changes in appetite, the proportion of participants on both arms who were actively smoking while on study was determined. In the hydrazine group, 19 of 32 patients (61%) were smoking compared with 21 of 33 (64%) in the placebo group. No patients in either hydrazine or placebo arm received long-term corticosteroid therapy. Antiemetic regimens involving short-term corticosteroid (usually one dexamethasone dose) were somewhat more common in patients receiving placebo (76%) compared with hydrazine (53%), but the difference was not statistically significant.
Hydrazine sulfate addition was not associated with change in objective tumor response frequency, which was comparable in both treatment arms for patients with measurable disease (partial response frequency, 23% for hydrazine v 29% for placebo groups, respectively, NS). No complete responses were seen. Allocation to hydrazine or placebo treatment also did not influence chemotherapy drug delivery, which was nearly optimal, following protocol standards with patients receiving close to 100% of protocol mandated dosage during the two initial treatment cycles (98.7 ± 3. 1% for hydrazine v 101.3 ± 2.8% for placebo, NS). Details of chemotherapeutic drug delivery during the first two treatment cycles are provided in Table 5. Radiotherapy given for disease progression on clinical indication "off study" was also closely comparable (occurring in six of 32 [19%] hydrazine v eight of 33 [23%] placebo patients, NS).
Hematologic toxicity was comparable in the two groups with no episodes of life-threatening myelosuppression seen. Similarly, gastrointestinal toxicity was as expected from a cisplatin-containing chemotherapy regimen. However, three patients could not continue the daily oral hydrazine sulfate treatment based on additional nausea. The proportion of patients who experienced no nausea was somewhat higher in the group receiving placebo versus hydrazine addition to chemotherapy (47% v 33%, respectively, NS).
In a randomized trial, the influence of hydrazine sulfate versus placebo addition to chemotherapy on clinical outcome was compared in patients with unresectable NSCLC. Overall survival was somewhat greater in the hydrazine group, and in the subset of fully ambulatory patients, a statistically significant increase in median survival was associated with hydrazine addition to chemotherapy.
Hydrazine sulfate also improved nutritional parameters in patients with NSCLC. The significant degree of albumin maintenance associated with hydrazine provides biologic plausibility for the favorable survival result, since this parameter is prognostic of long-term survival in lung cancer. In an analysis of 893 NSCLC patients with unresectable disease, maintenance of albumin level at 3 months from chemotherapy initiation was one of two factors significantly predictive of 2-year survival in the ECOG experience.
Up to the present time, chemotherapy alone has achieved a moderate impact on survival in NSCLC. The four randomized studies that compared cisplatin-containing combination chemotherapy to supportive care for advanced unresectable NSCLC patients are outlined in Table 6. Both the response frequency of 29% and the median survival of 28 weeks (197 days) seen in the current trial for patients in the chemotherapy plus placebo arm are closely comparable to results with chemotherapy reported by those investigators. Compared with such results with chemotherapy alone, the enhancement of chemotherapy impact seen with hydrazine represents a potential advance in NSCLC that will require confirmation in subsequent study. The moderate size of the current trial precludes definitive assessment of the clinical efficacy of hydrazine sulfate but does provide support for multiinstitutional study of this agent in NSCLC therapeutic regimens.
This trial was designed to evaluate the potential biological activity of hydrazine sulfate in NSCLC. The sample size was prospectively determined, based on the metabolic effects of hydrazine sulfate seen in pilot study. Although factors recognized, as being of prognostic value were relatively well balanced in both arms, multivariate analyses identified PS, weight loss, and liver involvement as the final variables with statistical significance.
Only patients with relatively favorable PS (PS 0-1) demonstrated statistically significant benefit on survival from hydrazine sulfate in this trial. The suggestion that survival benefit is limited to the PS 0-1 group is in agreement with other clinical trials involving chemotherapy in patients with NSCLC .
We have previously reported that hydrazine sulfate reduces both fasting glucose productions and lysine flux resulting in albumin maintenance in cancer patients. Enteral feeding also may reverse metabolic abnormalities associated with cancer cachexia. Whether reversal of abnormal host metabolism is the mechanism for hydrazine sulfate clinical effect will require future trials assessing both metabolic parameters and clinical outcome. Since hydrazine has modest cytotoxic activity and has occasional sedative effects, a mechanism involving either direct influence on tumor growth or one mediated centrally cannot be excluded. Nonetheless, the prospect that cancer patient survival may be increased by improving host metabolism represents a fundamentally new direction for clinical cancer management.
Hydrazine sulfate has not previously been prospectively evaluated in untreated populations with a single type of cancer using a randomized' study design. Hydrazine evaluation has been limited to single-arm, noncomparative reports involving heavily pretreated cancer populations. Such trials, which used subjective therapeutic end points, have largely reported positive results, whereas small trials of hydrazine in heavily treated populations, where reduction in tumor size was used as therapeutic end point, have reported little benefit. Our current results, indicating potential impact on nutritional and clinical end points, provide an impetus to further evaluation of hydrazine sulfate as an adjunct to conventional cancer therapeutic approaches in patients with NSCLC.
This page is designed and hosted by
Generation Computer Systems,
and is the property of the Syracuse Cancer
Research Institute. ©
1996, Syracuse Cancer Research Institute. All rights