Long-term results of blood and marrow transplantation for Hodgkin's lymphoma.

PURPOSE: To evaluate the long-term outcome after allogeneic (allo) and autologous (auto) blood or marrow transplantation (BMT) in patients with relapsed or refractory Hodgkin's lymphoma (HL). PATIENTS AND METHODS: We analyzed the outcome of 157 consecutive patients with relapsed or refractory HL, who underwent BMT between March 1985 and April 1998. Patients

Fludarabine and cyclophosphamide with filgrastim support in patients with previously untreated indolent lymphoid malignancies.

To evaluate the response rate and potential toxicities, a phase II trial was conducted of fludarabine and cyclophosphamide with filgrastim support in patients with previously untreated low-grade and select intermediate-grade lymphoid malignancies. Symptomatic patients with preserved end organ function received cyclophosphamide 600 mg/m(2) intravenous (iv) day 1 and fludarabine 20 mg/m(2) iv days1 through 5, followed by filgrastim 5 microg/kg subcutaneous starting approximately day 8. Treatment was repeated every 28 days until maximum response or a maximum of 6 cycles. Sixty patients, median age 53.5 years, were enrolled. Thirty-seven patients with non-Hodgkin lymphoma (NHL) were stage IV and 6 were stage III. Eleven of 17 patients with chronic lymphocytic leukemia (CLL) were Rai intermediate risk and 6 were high risk. The overall complete response (CR) rate was 51% and the partial response (PR) rate was 41%. Of patients with CLL, 47% achieved a CR and the remaining 53% achieved a PR. Of patients with follicular lymphoma, 60% achieved CR and 32% achieved a PR. Although the toxicity of this regimen was mainly hematologic, significant nonhematologic toxicities, including infections, were seen. Twenty-four patients subsequently received an autologous or allogeneic stem cell transplant. Engraftment was rapid, and there were no noticeable procedure toxicities in the immediate posttransplant period attributable to the fludarabine and cyclophosphamide regimen. Fludarabine, cyclophosphamide, and filgrastim make up a highly active and well-tolerated regimen in CLL and NHL.

Immunoablative high-dose cyclophosphamide without stem-cell rescue for refractory, severe autoimmune disease.

BACKGROUND: Immunoablative high-dose cyclophosphamide without stem-cell rescue induces durable, complete remission in most patients with aplastic anemia. OBJECTIVE: To determine the efficacy of high-dose cyclophosphamide in various refractory, severe autoimmune diseases. DESIGN: Prospective phase II study. SETTING: Johns Hopkins University (Baltimore, Maryland) and Hahnemann University (Philadelphia, Pennsylvania). PATIENTS: Eight patients with refractory, severe autoimmune disease. INTERVENTION: Immunoablative high-dose cyclophosphamide (50 mg/kg of body weight per day) for 4 consecutive days. MEASUREMENTS: Clinical and laboratory variables of autoimmune disease. RESULTS: Seven patients improved markedly: Five achieved complete remission and two achieved partial remission. Four patients have remained in continuous complete remission for 3 to 21 months, and two patients in partial remission continue to improve after 14 and 19 months of follow-up. High-dose cyclophosphamide was well tolerated; median times to a neutrophil count of 0.5 x 10(9) cells/L and platelet transfusion independence were 17 and 16 days, respectively. CONCLUSIONS: Immunoablative high-dose cyclophosphamide without stem-cell rescue can induce complete remission in patients with refractory, severe autoimmune disease. Reemergence of marrow function is similar to that seen after autologous transplantation and does not carry the risk for reinfusion of autoaggressive lymphocytes with the autograft.

Time-dependent effect of glutaraldehyde on the tendency to calcify of both autografts and xenografts.

To determine mechanisms responsible for the reduced calcification in short-term glutaraldehyde (Glu)-treated autologous pericardial bioprostheses, we studied the time effect of Glu on subsequent calcification and differences in calcification of autograft and xenograft implants in a rat subcutaneous implantation model. In experiment 1, four groups of bovine pericardial pieces (1 cm2) were prepared: (A) fresh bovine pericardium without Glu, (B) with 15-minute Glu, (C) with 60-minute Glu, and (D) with 120-minute Glu. Seven young male Sprague-Dawley rats were used; each received four bovine pericardial pieces from group A, B, C, or D for subcutaneous implantation. Calcium content of the implants (microgram/mg dry weight) 45 days later was 4.8 +/- 2.9, 29.8 +/- 13.6, 106.3 +/- 13.7, and 176.3 +/- 85.5 in groups A, B, C, and D, respectively (p < 0.05 between any two groups). Experiment 2 used 8 young male Sprague-Dawley rats from different mothers. Each received five subcutaneous skin implants. The five skin implants were prepared as follows: (1) fresh self skin, (2) self skin with 30-minute Glu, (3) self skin with 48-hour Glu, (4) fresh skin of others, and (5) skin of others with 48-hour Glu. After 45 days of implantation, the calcium content of the implants was 1.4 +/- 1.1, 57.9 +/- 35.4, 142.7 +/- 61.4, 1.5 +/- 1.1, and 94.9 +/- 24.1 micrograms/mg dry weight in groups 1, 2, 3, 4, and 5, respectively (p < 0.05 for 1 versus 2, 3, or 5; 2 versus 3, 4, or 5; 3 versus 4; and 4 versus 5).(ABSTRACT TRUNCATED AT 250 WORDS)

Intravascular lymphomatosis: a clinicopathologic study of 10 cases and assessment of response to chemotherapy.

PURPOSE: We report a clinicopathologic study of 10 cases of intravascular lymphomatosis (IVL) seen at a single institution, and assess the response to chemotherapy in these patients, as well as those collected from a literature review. PATIENTS AND METHODS: The clinical, pathologic, and immunophenotypic features of 10 cases of IVL diagnosed at the Johns Hopkins Hospital since 1977 were studied. Follow-up information was obtained in each case by consultation with the treating physician. In addition, cases of IVL reported previously in which patients were treated with chemotherapy and for which follow-up data were available at the time of publication were reviewed. RESULTS: In the present series of 10 cases, the most common clinical features were fever of unknown origin (FUO), mental status changes, and rash. Diagnostic specimens were obtained from a variety of sources, including brain, skin, prostate, liver, kidney, and gallbladder. All of the four patients treated with combination chemotherapy are alive and two have achieved long-term survival (48 and 45 months, respectively); the remaining two are alive and in complete remission (CR) after short follow-up duration of 6 months. Among 35 patients reported in the literature who received chemotherapy (including four from this series), 43% attained a CR and were free of disease at the time of publication. None of the three patients in our series who received localized therapy (surgery with or without radiation therapy) is alive (mean survival duration, 9 months). For the three patients diagnosed at postmortem examination, the mean interval between onset of symptoms and death was 3 months, and disease was widespread. CONCLUSION: These findings suggest that IVL represents a high-grade non-Hodgkin's lymphoma (NHL) with a propensity for systemic dissemination, and that CR and long-term survival may result in patients treated with aggressive combination chemotherapy.

Bioprosthetic cardiac valve degeneration: role of inflammatory and immune reactions.

Inflammatory and immune reactions are thought to mediate both calcification and biodegradation of bioprosthetic cardiac valve implants. To investigate the mechanisms of implant degeneration, we evaluated the role of inflammatory and immune reactions and the effects of tissue preservative treatment in three series of experiments. In the first experiment, three kinds of implants, i.e. glutaraldehyde-treated autograft Sprague-Dawley (SD) rat skin, xenograft Swiss-Webster (SW) mouse skin, and saline-treated autograft (control) were subcutaneously implanted in ten weanling SD rats, and retrieved after 70 days. There was no significant difference in the level of calcification in the autograft (113.13 +/- 27.09 micrograms/mg dry weight) and xenograft (78.27 +/- 31.53 micrograms/mg dry weight) (p > 0.05), but both differed significantly from the control specimens (1.55 +/- 0.87 micrograms/mg dry weight). In the second experiment, the immunological response to glutaraldehyde-treated bovine pericardium (glut tBP) and glycerol treated bovine pericardium (glyc tBP) implants were tested in vivo and in vitro. A Gore-Tex implant was used as a control. Sections of these materials were implanted to the abdominal muscle wall of Lewis rats, with each group composed of twelve animals. Lymphocytes and sera from the animals were isolated, and histological examination was performed at two or four weeks post-implantation. Collagen type 1 (calf skin) was used as antigen. Tritiated thymidine incorporation was used to measure lymphocyte response to antigen collagen type 1 (calf skin), and an Enzyme Linked Immunosorbent Assay (ELISA) was used to test antibodies. The results showed that lymphocytes from both the glut tBP and the glyc tBP groups responded to collagen type 1. The ELISA results showed that the glyc tBP group produced more antibodies than did the glut tBP group, with the difference being significant at a level of p < 0.02. Histology revealed that the glyc tBP had greater inflammatory changes and collagen degeneration than did the glut tBP. In the third experiment, sections of glut tBP and glyc tBP were implanted subcutaneously in two groups of ten weanling SD rats, and retrieved after 70 days. The results showed that glut tBP caused more calcification (197.04 +/- 83.56 micrograms/mg dry weight) than did the glyc tBP (6.74 +/- 0.55 microgram/mg dry weight), with the difference being significant at a level of p < 0.05. From these investigations it is concluded that tissue treatment prior to implantation was very important in determining the tendency of tissue to calcify, and that there was no obvious relationship between bioprosthetic calcification and immunogenicity.

Spontaneous host endothelial growth on bioprosthetic valves and its relation to calcification.

We studied host endothelial growth and calcification of bovine pericardial valve prostheses treated with: (A) 0.625% glutaraldehyde + 4% formaldehyde, (B) 99.5% glycerol or (C) 99.5% glycerol + 4% formaldehyde. Twenty-three stentless chordally supported bileaflet pericardial mitral valves with treatments A (n = 6), B (n = 6) or C (n = 11) were implanted in juvenile sheep for 125-273 days. After sacrifice, the anterior cusp from the annulus to papillary muscle of each valve was examined by scanning electron microscopy for the presence of endothelial cells, and the intrinsic calcification of each valve was determined by measuring calcium (micrograms/mg dry weight) from another 1 cm2 piece of grossly normal cusp. Sixty pieces of 1 cm2 pericardium with treatment A, B or C (n = 20 in each group) were implanted in 30 rats for 70 days. Calcium analysis and histology study of the implants were performed. In sheep, within a similar range of implantation periods, the endothelial growth rate of the valves was the highest in group B, 100% (6/6); group C was 45.5% (5/11) and A 16.7% (1/6). There were no significant differences in calcium among groups A, B and C. In rat implants, the calcium of group B was much lower than that of A or C (B = 6.92 +/- 4.46 vs A = 144.52 +/- 27.66 or C = 240.54 +/- 13.47, P < 0.05) although its histology showed more severe degeneration and inflammatory changes. Pericardial mitral valves treated with glycerol show satisfactory biocompatibility with regard to host endothelial growth and prevention of calcification; however, these tissues show evidence of rapid degeneration.(ABSTRACT TRUNCATED AT 250 WORDS)

Diphenylhydantoin inhibits calcification of bovine pericardial implants and myocardium: a preliminary study.

Calcification is a major cause of glutaraldehyde-fixed bioprosthetic valve failure. Recent studies have shown that dystrophic calcification shares basic features with normal bone mineralization, including crystal initiation through the mediation of cell membranes, usually in the form of extracellular vesicles. In this study, we observed that calcification of the myocardium of DBA/2J mice was inhibited or reversed by diets supplemented with 100 mg/kg diet diphenylhydantoin (dilantin) for 70 days, with a calcification incidence of 25% in the dilantin group versus 58% in control. We further studied the effects of dilantin on bioprosthetic valve calcification. Three groups of young male Sprague-Dawley rats (100 g, 9/group) were implanted subcutaneously with 1-cm2 pieces of glutaraldehyde-fixed bovine pericardium. Controls were fed a ground chow for 45 or 90 days postimplantation; experimentals received the same chow for the first 45 days postimplantation and then were fed the same diet supplemented with 1000 mg dilantin/kg for the succeeding 45 days. Calcium content (microgram/mg dry weight) of the implants in the dilantin group was 137 +/- 18.6 versus 214 +/- 34.3 in 90 days control and 79.9 +/- 41.5 in 45 days control (mean +/- SD, P < 0.01 and P < 0.05 respectively, t test). The tibia calcium content of the dilantin group was not significantly different from 90 days control. We conclude that orally administered dilantin inhibits calcification of glutaraldehyde-fixed bovine pericardial implants preferentially. It does not cause decalcification either of implants that have already calcified or of the bones. The anti-calcification effect of dilantin may be associated with its anti-vitamin D effect.

Bovine pericardium versus porcine aortic valve: comparison of tissue biological properties as prosthetic valves.

The choice of a bioprosthetic valve substitute remains controversial with the major concern being primary tissue failure after implantation. We compared biological properties of the two most frequently used bioprosthetic valve materials, bovine pericardium and porcine aortic valve, before and 90 days after subcutaneous implantation in rats. Before implantation, tissue collagen and water content were measured in nine pieces of bovine pericardium and porcine valves, each fixed in 0.625% glutaraldehyde; calcium, tissue collagen, and water content were measured in another nine pieces of the same tissues after 90 days' implantation. Bovine pericardium had higher collagen content than that of porcine valve (hydroxyproline, 7.98 +/- 0.05* vs. 4.56 +/- 0.02 micrograms/mg, dry weight) but lesser water content (72.16 +/- 3.22%* vs. 87.36 +/- 1.62%) before implantation (*p < 0.001, mean +/- SD, t test); after implantation, bovine pericardium still maintained higher collagen content (hydroxyproline, 4.89 +/- 0.04* vs. 2.61 +/- 0.06 micrograms/mg, dry weight) but contained the same amount of water (60.24 +/- 5.08% vs. 61.43 +/- 9.00%) and calcium (214.43 +/- 34.34 vs. 199.33 +/- 53.44 micrograms/mg, dry weight) (*p < 0.001, mean +/- SD, t test). We conclude that bovine pericardium has superior intrinsic biological properties for prosthetic valve manufacture. With proper integration of properties and design it will in some applications be superior to the porcine aortic valve.

The role of aromatic hydrocarbons in the genesis of breast cancer.

The incidence of breast cancer in women has increased dramatically over the last decade. Epidemiological markers of this increased incidence include: endocrine related phenomena (early menarche, age of first parity and age of menopause); exposure of the breast to X-radiation; and a group of seemingly disparate factors--urban residence, dietary selection and alcohol consumption. Although experimental breast cancer may be induced by estrogenic hormones, X-radiation and aromatic hydrocarbons, only aromatic hydrocarbons have not been previously implicated in human mammary carcinogenesis. The seemingly unrelated human factors can best be understood by examining the role of breast tissue in aromatic hydrocarbon metabolism. Aromatic hydrocarbons are important environmental chemicals produced by the incomplete combustion of hydrocarbons for use in energy production. Benzene, benz(a)pyrene, dibenz(ah)anthracene and 1-nitropyrene, known experimental breast carcinogens, are produced in this way. Human exposure to aromatic hydrocarbon metabolites induces and promotes altered DNA by mechanisms described as increased intracellular pro-oxidant production as well as direct adduction to DNA. The breast is anatomically embedded in a major fat depot which stores and concentrates aromatic hydrocarbons and can metabolize these hydrocarbons to carcinogenic metabolites. Ductal cells concentrate these metabolites and themselves become target cells for carcinogenesis. Some lifestyle factors increase the amount of carcinogens produced or enhance their activity. A unitary model for mammary carcinogenesis in humans as well as in experimental carcinogenesis is hypothesized. If correct, the hypothesis would account for some of the increase in breast cancer incidence in industrial countries--and would suggest environmental and dietary modifications that would inhibit hydrocarbon induced mammary carcinogenesis.

Improved postfixation treatment of glutaraldehyde fixed porcine aortic valves by monosodium glutamate.

The use of bioprosthetic valves remains limited due to poor long-term durability primarily because of tissue calcification-associated degeneration. Release of locally cytotoxic residual aldehyde after glutaraldehyde fixation is one of the major causes of this degeneration. In this study, monosodium glutamate was used as postfixation treatment to bind residual aldehyde in order to block its toxic effects. Thirty-six pieces of fresh porcine aortic valves were fixed by 0.625% glutaraldehyde for 14 days, and then 18 of them were treated with 1% monosodium glutamate for another 3 days before they were implanted subcutaneously into the backs of two groups of rats (n = 9 in each group) for 45 and 90 days, respectively. Retrieved specimens were examined grossly, and calcium analysis and measurements of tissue collagen and water content were carried out. The results showed that, compared with glutaraldehyde fixed specimens, monosodium glutamate postfixation treated specimens had less calcification (calcium 104.93 + 50.94 versus 141.58 +/- 58.10 at 45 days and 103.07 +/- 76.48 versus 199.33 +/- 53.44 at 90 days, micrograms/mg dry weight, p < 0.01), higher collagen content (hydroxyproline 5.50 +/- 1.29 versus 3.58 +/- 1.48 at 45 days and 5.64 +/- 0.87 versus 4.25 +/- 0.65 at 90 days, micrograms/mg wet weight, p < 0.01), and higher water content (68.00 +/- 6.95% versus 61.33 +/- 8.83% at 90 days, p < 0.05) (mean +/- SD, paired t test). We conclude that monosodium glutamate couples with residual aldehyde, which significantly reduces calcification of glutaraldehyde fixed porcine aortic valves while preserving a higher tissue collagen and water content after implantation. The preserved tissue collagen and water content of the implants is closer to that of unimplanted native valves.

Improved biocompatibility of bovine pericardium using a new method of cross linking.

The authors used glycerol (Gly) as a cross linking agent to treat bovine pericardium (BP) and compared it mechanically (tensile test) and biologically (subcutaneous implantation in rats) with glutaraldehyde (Glu) fixed lonescu-Shiley BP (ISBP). Maximum tensile stress (Stmax) was the same in both groups (n = 12 each), while maximum strain (Snmax) was larger in Gly BP. Calcium content (Ca) was lower and water content of tissue (WC) was higher in implants of Gly BP versus ISBP (n = 9 each). Gly BP appeared more hydrated than ISBP in specimens studied by scanning electron microscopy (SEM). In addition, SEM of three Gly BP mitral valves in sheep showed good growth of endothelial (En) cells. These biomechanical advantages of Gly BP over ISBP suggest that Gly may be a useful substitute for Glu in cross linking bovine pericardial valves.

Aldehyde tanning: the villain in bioprosthetic calcification.

Preservation of bioprosthetic valves may play a role in valvular calcification. Subcutaneous implants in rats were used to test the effect of different preservation solutions. Fifty male Sprague-Dawley rats were divided into five groups. Fresh bovine pericardium was treated in one of five ways: group A: 99.5% glycerol for 1 week; group B: as group A, then normal saline wash and 0.25% formaldehyde storage for 24 h; group C: as group A, then normal saline wash and 4% formaldehyde storage for 24 h; group D: as group A, then normal saline wash and 0.625% glutaraldehyde storage for 24 h; group E: 0.625% glutaraldehyde and 4% buffered formaldehyde storage. Treated bovine pericardium was cut into 1-cm2 pieces and washed for 30 min with normal saline before implantation. In each animal, three pieces were implanted in the subcutaneous tissue of the back. After 70 days, retrieved specimens were examined grossly, and X-ray densitometry, calcium analysis, and histological examinations were carried out. The results showed that glycerol-treated tissue (group A) had less calcification (calcium 6.92 +/- 4.46 micrograms/mg dry weight) than other groups: group B (calcium 323.12 + 63.56 micrograms/mg dry weight); group C (calcium 240.65 + 13.47 micrograms/mg dry weight); group D (calcium 232.29 + 13.01 micrograms/mg dry weight). These differences were markedly significant (p less than 0.0001). It appears that aldehydes play an important role in the calcification of bioprosthetic valves. Experience with glutaraldehyde- and glycerol-treated pericardium in valvular applications in sheep support these observations.

A phase II trial of carboplatin (CBDCA) in small-cell and non-small-cell lung cancer with correlation to in vitro analysis of cytotoxicity.

A Phase II trial of carboplatin (CBDCA) was performed in 33 patients with advanced lung cancer, including 15 patients with inoperable Stage III non-small-cell (NSCLC) and 18 patients with relapsed small-cell (SCLC) lung cancer. Initial dosage was 320 mg/m2 infused over 24 h; in the absence of hematologic toxicity, subsequent doses were escalated to 400 mg/m2. Patients received a median of two cycles (range 1-13 for NSCLC and 1-5 for SCLC) of therapy. There were no complete or partial responses among the NSCLC patients. Among the SCLC patients, two had a partial response. In vitro analysis of the cytotoxicity of CBDCA and its parent compound cisplatin by two different methods for 20 NSCLC cell lines suggested that equivalent tumor cell kill is achieved by the two compounds, but this occurs at a log lower concentration of cisplatin than of CBDCA. The in vitro cytotoxicity against NSCLC of CBDCA at a concentration predicted to be in the range produced by the dose employed in this Phase II study correlated well with the resulting very modest in vivo benefit. In vitro, a continuous dose-response relationship exists for CBDCA, suggesting that if higher doses could be administered safely to patients, greater clinical benefit might occur. We conclude that single agent CBDCA in the dosage and schedule administered has less than 20% activity (95% confidence intervals 0-19%) in NSCLC and an 11% response rate in SCLC (95% confidence intervals 2-34%). Despite this outcome, in vitro data in human NSCLC cell lines suggest higher dosages should perhaps be evaluated before discounting a role for CBDCA in the management of NSCLC.

Zone II flexor tendon repair: effects of vitamins A, E, beta-carotene.

Ninety-six adult Leghorn chickens each had the flexor profundus tendon in each middle toe sharply divided in Zone II with immediate repair (pentobarbital, ketamine anesthesia). Animals were then randomly assigned to receive unsupplemented standard chick chow or the chow supplemented with vitamin A (150,000 IU/kg chow), Vitamin E (1000 IU/kg chow), or beta-carotene (90 mg/kg chow). Eight animals from each of the four groups were examined at 7, 30, or 45 days post repair. After sacrifice, in situ composite wound breaking strength was measured in the amputated toe by constant speed tensiometry. Vitamin A-supplemented animals demonstrated breaking strength more than double that of control at each postoperative test day, while those animals receiving supplemental Vitamin E had breaking strength less than half that of control at Day 7 and Day 45. These results are statistically significant. Tensiometry curves differed markedly at all time points among the groups: Vitamin A curves being broader, higher, and having more spikes. These differences in the tensiometry curves, both qualitative and quantitative, may be due to differences in intrinsic tendon healing or to differences in adhesion formation or a combination of both. beta-Carotene supplementation had modest effect. We conclude that supplemental dietary vitamin A increases the breaking strength of composite tendon wounds and that supplemental dietary vitamin E decreases it.

High-dose cytotoxic therapy and bone marrow transplantation for relapsed Hodgkin's disease.

Patients with Hodgkin's disease who have failed two or more chemotherapy regimens or who have relapsed after an initial chemotherapy-induced remission of less than 12 months are seldom cured with conventional salvage therapies. We studied the effect of high-dose cytoreductive therapy followed by bone marrow transplantation in 50 such patients with relapsed Hodgkin's disease. Twenty-one patients with histocompatibility locus antigen (HLA)-matched donors had allogeneic marrow transplants, one patient received marrow from an identical twin, and 28 patients without a matched donor received autologous grafts purged with 4-hydroperoxycyclophosphamide. Busulfan plus cyclophosphamide was the preparative regimen for the 25 patients who had received extensive prior irradiation, and the other 25 patients received cyclophosphamide plus total body irradiation. The overall actuarial probability of event-free survival at 3 years was 30%, with a median follow-up of 26 months. The event-free survival following transplantation was influenced by the number of chemotherapy failures and the patient's response to conventional salvage therapy prior to transplant. The 16 patients who were transplanted at first relapse, while still responsive to standard therapy, had a 64% actuarial probability of event-free survival at 3 years. Age, presence of extranodal disease, preparative regimen, and type of graft (autologous v allogeneic) were not significant prognostic factors. The majority of transplant-related deaths were from interstitial pneumonitis; inadequate pulmonary function, multiple prior chemotherapy regimens, and prior chest irradiation all appeared to increase the transplant-related mortality. These results suggest a role for marrow transplantation in a subset of patients with relapsed Hodgkin's disease who are unlikely to be otherwise cured but are still responsive to conventional-dose cytoreductive therapy.

Supplemental vitamin A prevents the tumor-induced defect in wound healing.

To test our hypothesis that supplemental vitamin A would mitigate the impaired healing that occurs in tumor-bearing animals, six groups of C3H mice, eight per group, eating a standard commercial mouse chow ad libitum that supports normal growth, reproduction, and longevity were innoculated with 200,000 C3HBA cells. When tumors measured approximately 6 mm in diameter, the mice were anesthesized and wounded (dorsal skin incisions and subcutaneous polyvinyl alcohol sponges). Twenty-four hours later, two groups (one continued on the chow and the other started on the chow supplemented with 150,000 IU vitamin A/kg chow) underwent local tumor irradiation; two groups, one ingesting the chow, the other the vitamin A supplemented chow, were started on cyclophosphamide therapy; two groups, one ingesting the chow, the other the vitamin A supplemented chow, received neither local tumor irradiation nor cyclophosphamide therapy. An additional two groups ingesting the chow, one group neither innoculated with tumor nor wounded, the other wounded by not innoculated, served as controls. Wound breaking strength and sponge reparative collagen accumulation (assessed by hydroxyproline proline measurement) were used as indicators of wound healing. The mice were killed 12 days after wounding. Tumor presence decreased wound breaking strength and sponge hydroxyproline content; these effects were largely negated by supplemental vitamin A. Local tumor irradiation diminished the adverse effect of tumor on sponge reparative collagen content but to a lesser extent than the supplemental vitamin A. Supplemental vitamin A added to the irradiation effect on healing but irradiation did not add to the vitamin A effect. Cyclophosphamide, a systemic radiomimetic anti-tumor agent, did not alter the impaired wound healing of the tumor-bearing mice. Supplemental vitamin A mitigated the impaired wound healing in the cyclophosphamide-treated tumor-bearing mice. Supplemental vitamin A also moderated the effects of wounding, tumor, and tumor therapies (local irradiation and cyclophosphamide) on the increase in adrenal size, leukopenia, thrombocytopenia, and thymic involution (except the last was not moderated in the cyclophosphamide-treated tumor-bearing rats). The splenic enlargement in the untreated tumor-bearing wounded rats and in those treated with cyclophosphamide was lessened by supplemental vitamin A. We hypothesize that these anti-stress effects of vitamin A underlie, in part, its action in mitigating the impaired wound healing of tumor-bearing mice, including those treated by local irradiation or cyclophosphamide. These findings have implications for the care of patients with malignant tumors.