Elevated c-myc messenger RNA in multiple myeloma cell lines.

Oncogene analyses of four human myeloma cell lines provided no indication of gene amplification or rearrangement using DNA probes for the met, raf, abl, mos, erb B, Her-2-neu, fos, myb-7, fms, L-myc, sis, and myb-1 genes. However, a consistent elevation of up to 23-fold in the level of c-myc mRNA was observed in all of the cell lines studied. No restriction fragment length polymorphism (in exons one, two, or three) or c-myc gene amplification has as yet been demonstrated to account for the c-myc mRNA elevation. The c-myc mRNA has a half-life of 25 min which is comparable to that observed in other systems. The elevation in c-myc mRNA is further evidence for the role of the c-myc proto-oncogene in the pathogenesis of myeloma.

Reassessment of the relationship between M-protein decrement and survival in multiple myeloma.

The relationship between percentage M-protein decrement and survival is assessed in 134 multiple myeloma patients. The correlation did not achieve statistical significance (P = 0.069). Multivariate analysis using the Cox proportional hazards model, including a number of previously recognised prognostic factors, showed only percentage M-protein decrement, creatinine and haemoglobin to be significantly correlated with survival. However, the R'-statistic for each of these variables was low, indicating that their prognostic power is weak. We conclude that neither the percentage M-protein decrement nor the response derived from it can be used as an accurate means of assessing the efficacy of treatment in myeloma. Mature survival data alone should be used for this purpose.

Dose intensity analysis of melphalan and prednisone in multiple myeloma.

The average relative dose intensity (DI) of conventional oral melphalan and prednisone therapy received by 93 newly diagnosed multiple myeloma patients was correlated with survival and with percent reduction in M-protein. A survival advantage was shown with increasing average relative DI of melphalan and prednisone. Multivariate analysis showed survival to correlate with increasing DI of prednisone (P = .05) but not with the DI of melphalan (P = .93) nor with the percent decrement in M-protein (P = .10). These results suggest that the initial management of myeloma should be reassessed, with particular emphasis on more intensive therapy employing high-dose steroids.

Are the current criteria for response useful in the management of multiple myeloma?

One hundred seventy-three patients with multiple myeloma were treated from the time of diagnosis with standard oral melphalan and prednisone at 28-day intervals until they became refractory to treatment. Response to treatment was determined according to the Chronic Leukemia-Myeloma Task Force (TF) criteria, and independently according to the Southwest Oncology Group (SWOG) criteria. Survival by disease stage and response according to the two sets of criteria were analyzed for patients living longer than 3 months. The median survival of responding and nonresponding (TF criteria) stage II patients was 43.8 and 40.3 months, respectively (P = .29). By SWOG criteria, median survival for responding and nonresponding stage II patients was 48.3 and 39.0 months, respectively (P = .12). In stage III patients, median survival for responders and nonresponders (TF criteria) was 34.0 and 21.7 months, respectively (P = .01), compared with 35.5 and 24.4 months (P = .04) by SWOG criteria. These data would suggest that the TF criteria predicts a survival disadvantage only in very advanced myeloma and that applying the stricter limits for the definition of response of the SWOG does not further aid in selecting a subgroup of myeloma patients with poorer survival.

Cytogenetic and biological characterization of two new human plasma cell lines.

Two new human plasma cell lines designated as ACB-885 and ACB-1085 have been established from a 39-year-old patient with multiple myeloma. These cell lines have definitive plasma cell features by morphologic examination, and essentially all of the cells are positive for cytoplasmic IgG kappa immunoglobulin. These cells are negative for standard T-cell surface markers and mature B-cell markers, such as B1, B2, and HLA-DR, but are strongly positive for the antigen defined by OKT-10. The cells are negative for Epstein-Barr virus. The cell lines have a doubling time of 30-35 hours and a growth fraction approaching 100%. Cytogenetic analysis showed a 2n chromosome number of 45-46 with very similar karyotypic abnormalities in both the plasma cell lines and the original tumor material. One of the chromosomes in each of the pairs of chromosomes number #1, #2, #6, #7, #8, #10, #12, #13, and #22 were consistently missing. These were replaced by eight marker chromosomes that resulted from chromosomal rearrangements involving mainly these missing chromosomes. Almost all of the breakpoints occurring in the marker chromosomes were identified, and eight of these breakpoints have been reported in other studies of myeloma plasma cells. Homogeneously staining regions were observed in two marker chromosomes suggesting gene amplification in these chromosomal regions.

Loss of viability and induction of DNA damage in human leukemic myeloblasts and lymphocytes by m-AMSA.

The effects of m-AMSA on in vitro viability and on the induction of DNA damage were examined in low-growth-fraction cell populations of human leukemic myeloblasts and normal lymphocytes. A significant individual variation in the drug-induced reduction of in vitro viability was observed in studies with five selected leukemic patients. The concentration of m-AMSA required to reduce viability by 50% within 48 h ranged from 0.25 microM to in excess of 5.0 microM for the leukemic myeloblasts as against about 2.0 microM for the samples of normal lymphocytes. Alkaline elution studies showed that m-AMSA induced protein-associated DNA strand breaks (PADB) in both myeloblasts and lymphocytes. Depending upon the m-AMSA concentration, there was a 4- to 9-fold difference in the level of PADBs induced by a given drug concentration in the myeloblasts of eight patients studied. The level of PADBs was saturable with respect to both drug concentration (5-10 microM) and exposure time (45-10 microM). The PADBs were repaired rapidly in all the lymphocyte and myeloblast samples studied, with over 90% of this DNA damage being repaired within 45 min after resuspension of the cells in drug-free medium. These studies of m-AMSA in low-growth-fraction samples of human lymphocytes and myeloblasts show both similarities and differences in the action of this drug compared with previously published studies using the high-growth-fraction mouse L1210 system.

Single-strand breaks or alkali-sensitive sites in the DNA of human myeloma plasma cells and chronic lymphocytic leukemia lymphocytes.

Alkaline-elution studies showed significant levels of either DNA single-strand breaks or alkali-sensitive sites in the plasma cells of six out of six myeloma patients and in the lymphocytes of two out of four patients with chronic lymphocytic leukemia as compared with normal human lymphocytes. The increased rate of DNA elution was variable from sample to sample with a range that would correspond to that observed with 100-1000 rad (1 rad = 10 mGy) of X-ray irradiation. This alteration in DNA structure was observed in both new and advanced patients, did not appear to be related to prior therapy, and did not affect the in vitro viability of these cells. Repetitive alkaline-elution profiles obtained with tumor cells from three patients were similar on subsequent samples obtained 1 month apart. Altered DNA elution was not evident in peripheral blood lymphocytes from myeloma patients with altered plasma cell DNA elution. These observations are interesting in light of the recent hypothesis that breaking and rejoining of DNA, regulated by poly(ADP-ribosyl)ation, may be a general mechanism of altering gene expression during differentiation.

Treatment of multiple myeloma with deoxycoformycin.

A comparison of adenosine deaminase activity in intact human plasma cells and lymphocytes in vitro showed that plasma cells had at least as much activity of this enzyme as did T or non-T lymphocytes. This observation led us to examine the effectiveness of deoxycoformycin in the treatment of multiple myeloma. Thirteen patients with advanced refractory myeloma were treated with deoxycoformycin at 5 mg/m2 daily for 3 days every 2 weeks until response or progression. Of the seven evaluable patients who received more than one cycle of therapy, two had a greater than 50% reduction in the level of myeloma protein and two had a demonstrable reduction in soft tissue disease. Toxicity consisted of marked nausea, anorexia lasting several days, and mild transient confusion in some patients. Plasma levels of deoxyadenosine and adenosine peaked on day 4 or 5 with average values of 1.9 and 0.6 microM, respectively. Red cell levels of dATP reached approximately 40% of ATP levels. The viability of plasma cells was shown to be greatly reduced in in vitro incubations with deoxycoformycin and low levels of deoxyadenosine (ID50 of 6 microM).

Aphidicolin and deoxycoformycin cause DNA breaks and cell death in unstimulated human lymphocytes.

Human lymphocytes lose viability when incubated in vitro with either aphidicolin, an inhibitor of DNA polymerase alpha, or with the combination of aphidicolin and deoxycoformycin (an adenosine deaminase inhibitor). Loss of viability was assayed by vital staining with fluorescein diacetate as well as examination of Wright stained preparations and the appearance of cellular debris observed using an electronic cell counter. The loss of viability was rapid with the combination of aphidicolin (2 micrograms/ml) and deoxycoformycin (1 microgram/ml) with essentially complete loss of viability after 72 hours of incubation. This drug combination produces DNA single strand breaks after 24 and 48 hours of incubation at a level equivalent to that produced by 200 or 400R of X-irradiation, respectively.

DNA strand breaks induced in human T-lymphocytes by the combination of deoxyadenosine and deoxycoformycin.

There is a progressive loss of human T-lymphocyte viability upon incubation with deoxycoformycin, an adenosine deaminase inhibitor, and low concentrations of deoxyadenosine (drug concentration that reduced cell count at 48 hr after initiation to 50% of value for untreated control culture, less than 1 microM). The loss of viability was evidenced by vital staining with fluorescein diacetate and by changes in forward single light scatter measured by flow cytometry. This loss of lymphocyte viability is detectable 18 to 20 hr after the addition of deoxyadenosine and is earlier than has been reported by other investigators using trypan blue as the vital stain. Alkaline elution studies show that the incubation of T-lymphocytes with the combinations of deoxycoformycin and deoxyadenosine gives rise to DNA single-strand breaks. These DNA strand breaks are dose and time dependent and are readily detected 4 hr after the addition of deoxyadenosine. These DNA lesions are not observed with deoxycoformycin or deoxyadenosine alone. Incubations of T-lymphocytes with deoxycoformycin and deoxyadenosine (1 and 5 microM) for 7 hr result in DNA strand breaks with a frequency of 145 and 280 rad equivalents, respectively. Preliminary studies indicate that the ability of lymphocytes to repair this damage is dependent upon deoxyadenosine concentration and exposure time. The relationship of these DNA lesions to loss of lymphocyte viability in the presence of deoxycoformycin and deoxyadenosine remains to be established.

The effect of anoxia on anthracycline-induced DNA damage in the RPMI-6410 human lymphoblastoid cell line.

The alkaline elution procedure developed by Kohn and co-workers was used with the RPMI-6410 cultured human lymphoblastoid cell line to examine the hypothesis that anthracycline-induced DNA strand scission is mediated by oxygen- or superoxide-derived free radicals. Hypoxia was induced by gassing with nitrogen containing 5% carbon dioxide and less than 4 ppm oxygen. Alkaline elution studies showed hypoxia was induced, as the oxygen enhancement ratios for DNA strand breaks was 2.4 and 2.6 for the 250 R +/- oxygen and the 500 R +/- oxygen (1 R = 2.58 x 10(-4) C/kg) experiments, respectively. The pattern of adriamycin-induced DNA strand breaks and cross-linking was not affected by hypoxia with 1-h adriamycin exposures between 0.05 and 1.0 microgram/ml. Similarly, 1-h exposures of N-trifluoroacetyladriamycin-14-valerate at 3 or 10 micrograms/mL gave essentially identical alkaline elution profiles in the presence or absence of oxygen. These results do not support the hypothesis that oxygen-derived radicals play a primary role in anthracycline-induced DNA strand breakage.

Adenosine and deoxyadenosine toxicity in colony assay systems for human T-lymphocytes, B-lymphocytes, and granulocytes.

Adenosine and deoxyadenosine toxicity was examined in colony assay systems for human T lymphocytes, B lymphocytes, and granulocytes. In the absence of deoxycoformycin, an adenosine deaminase inhibitor, no growth inhibition was observed in the three systems with concentrations of adenosine or deoxyadenosine of at least 200 microM. Deoxycoformycin itself had no growth-inhibitory effect at concentrations of at least 10 micrograms/ml. Combinations of deoxycoformycin (1 microgram/ml) and either adenosine or deoxyadenosine gave growth inhibition in all three systems. Deoxyadenosine was the most toxic in all the systems, the LD50 values being 20-25 microM. The LD50 values for adenosine were 45-55 microM. There was no evidence of selective toxicity by adenosine or deoxyadenosine with these three colony assay systems. In the T-lymphocyte colony system deoxyadenosine appeared to be toxic to both the inducer/helper and the suppressor/cytotoxic T-lymphocyte subpopulations.

Effect of liver disease on pharmacokinetics and toxicity of 9-beta-D-arabinofuranosyladenine-5-phosphate.

Plasma levels of 9-beta-D-arabinofuranosyladenine-5'-phosphate (ara-AMP) and its metabolites 9-beta-D-arabinofuranosyladenine (ara-A and 9-beta-D-arabinofuranosylhypoxanthine (ara-Hx) were determined by high-performance liquid chromatography in four patients with chronic active hepatitis positive for hepatitis B surface antigen and eight patients with severe herpesvirus infections and normal liver function. Ara-AMP was given intravenously over a 30-min period in doses ranging from 10 to 30 mg of ara-A equivalent/kg per day. The metabolism of ara-AMP did not differ significantly between the two patient groups. Ara-AMP was quickly converted to ara-A, which was rapidly deaminated to ara-Hx. The mean half-lives of ara-AMP, ara-A, and ara-Hx were 0.14 hr, 0.17 hr, and 3.5 hr, respectively. Thus, ara-AMP is rapidly metabolized and does not act as a depot form of ara-A. Patients with chronic active hepatitis demonstrated increased bone marrow sensitivity to ara-AMP and a musculoskeletal pain syndrome not observed in patients treated for herpesvirus infections.

The DNA content of human plasma cells.

The DNA content of human plasma cells from myeloma patients relative to that of leukocytes was determined by flow and microscopic cytofluorometry after propidium iodide and fluorescent Feulgen staining, respectively. Mononucleated myeloma plasma cells from all of the 17 patients studied contained more DNA (17 to 58%) than the leukocytes from the patient. The binucleated and trinucleated plasma cells, which were more prevalent in advanced cases, contained up to two and three times, respectively, the amount of DNA determined in the mononucleated plasma cells. These observations suggest that the ploidy abnormalities of myeloma plasma cells are even more extensive than the numerous karyotypic studies have indicated.

Clinical pharmacology of oral thioguanine in acute myelogenous leukemia.

Although thioguanine has been in clinical use for over 20 years, few data are yet available on the clinical pharmacology of thioguanine administered orally. We have studied the plasma thioguanine levels in acute myelogenous leukemia patients during remission induction (daunomycin 60 mg/m2 on day 1, arabinosylcytosine 200 mg/m 2. day for 7 days by infusion, thioguanine 100 mg/m2 PO every 12 h for 7 days) and remission maintenance (arabinosylcytosine 200 mg/m2 . day for 4 days by infusion, thioguanine 100 mg/m2 PO every 12 h for 4 days). Hourly blood samples were taken after thioguanine administration, and plasma thioguanine levels were measured by high-performance liquid chromatography with an anion-exchange column. Prior to the chromatography the thioguanine was oxidized by alkaline potassium permanganate to the corresponding 6-sulfonate, which was monitored by means of fluorescence detection. Peak plasma levels of thioguanine were observed 2-4 h after administration and varied from 0.03-0.94 microM. Plasma levels of thioguanine were markedly lower in patients with severe nausea and emesis. Food intake at the same time as thioguanine administration also tended to lower plasma drug levels. The 30-fold range in thioguanine plasma levels observed in this study suggests that intermittent IV administration may provide a better means of standardizing the dosage of thioguanine.

N-trifluoroacetyladriamycin-14-valerate and adriamycin induced DNA damage in the RPMI-6410 human lymphoblastoid cell line.

The adriamycin- and N-trifluoroacetyladriamycin-14-valerate (AD-32) induced DNA cross-linking and breakage in human RPMI-6410 cells was compared using the alkaline elution technique of Kohn and co-workers. At comparable growth-inhibitory concentrations both adriamycin and AD-32 caused DNA cross-linking. Treatment with proteinase-K showed this cross-linking to be mainly DNA-protein in character. Proteinase-K treatment also revealed that both drugs caused either single-strand DNA breaks or increased alkaline sensitivity. With adriamycin the degree of cross-linking and breakage was dose related over the range studied (0.05 - 0.4 micron/mL), whereas with AD-32 there appeared to be a saturation of both effects at concentrations in excess of 3 micron/mL. With both drugs the extent of cross-linking and breakage was maximal at the end of the drug exposure. This work suggests that AD-32 or some metabolite of its binds to DNA and this binding leads to DNA damage that is similar to that caused by adriamycin. These AD-32 results are somewhat surprising in light of earlier model studies showing that AD-32 does not bind to isolated DNA.

L-phenylalanine mustard (melphalan) uptake and cross-linking in the RPMI 6410 human lymphoblastoid cell line.

L-Phenylalanine mustard (melphalan) induced a time- and concentration-dependent arrest of cycling RPMI 6410 cells in the G2 phase of the cell cycle as evidenced by flow cytofluorometry. A melphalan exposure of 1 microgram/ml for 1 hr caused a temporary G2 blockage which was overcome by 48 hr. Higher concentrations or longer exposures lead to irreversible blockages. Melphalan caused DNA cross-linking which was monitored by the alkaline elution method. The cross-linking was shown to be between DNA and protein. The degree of DNA cross-linking increased for approximately 4 hr after a 1-hr drug exposure of 1 microgram/ml. At 36 to 48 hr after the drug exposure, the cells overcame the G2 block and were dividing. The DNA cross-links have apparently been repaired as they are no longer detected by alkaline elution. The extent of melphalan cross-linking was dependent on both drug dosage and exposure time. Using a culture medium lacking amino acids, it was shown that melphalan uptake into RPMI 6410 cells was inhibited by leucine, isoleucine, or glutamine. The increased uptake of melphalan and the increased cross-linking in amino acid-deficient media were reduced by readdition of the aforementioned amino acids.