A phase I study of ifosfamide given on alternate days to treat children with brain tumors

BACKGROUND

Ifosfamide with Mesna, given every other day over a 5-day period, was evaluated in 20 children with recurrent or progressive primary brain tumors.

METHODS

The patients were assigned to dosage cohorts separated on the basis of prior exposure to cisplatin (n = 10) or the absence of such exposure (n = 10). The initial dose in each treatment arm was 2133 mg/m2 every other day for three doses, which represented 80% of the total dose delivered in our prior study of ifosfamide given daily over 5 days. The dose was escalated by 20% in each of the two subsequent cohorts (2560 mg/m2 and 3072 mg/m2 every other day for three doses).

RESULTS

The hematologic toxicity was dose limiting. Prior exposure to cisplatin did not seem to increase the hematologic toxicity. The most frequent and significant metabolic disturbance was hyponatremia, resulting in self-limited seizure activity in three patients. This complication was prevented in subsequent patients by changing the post-ifosfamide hydration fluids from 5% dextrose in quarter normal saline to 5% dextrose in normal saline.

CONCLUSIONS

Although no child achieved a complete response, the activity of ifosfamide was demonstrated for a variety of tumors. The recommended dose of ifosfamide in a Phase II study for brain tumors is 3000 mg/m2 given with Mesna every other day for three doses.

A phase I study of ifosfamide with Mesna given daily for 3 consecutive days to children with malignant solid tumors

BACKGROUND

The authors conducted a Phase I dose escalation trial of ifosfamide given daily for 3 consecutive days to 29 children with malignant solid tumors. Twenty-eight of these children had received prior chemotherapy.

METHODS

Patients were assigned to dosage cohorts separately on the basis of prior exposure to the platinum alkylating agents cisplatin or carboplatin (n = 20) or the absence of such exposure (n = 9). At least three patients in each category were treated at a starting dosage of 2133 mg/m2/d for 3 days. This dosage represented 80% of the total dose delivered in the prior study of ifosfamide given daily over 5 days with dosage escalation of 20% in subsequent cohorts.

RESULTS

Myelosuppression was dose-limiting at the second dosage level (2560 mg/m2/d) for patients previously treated with platinum and at the third dosage level (3072 mg/m2/d) for those not previously treated with platinum. Dose-limiting neurotoxicity was seen at 2560 mg/m2/d for the former group, but was not encountered in the latter group.

CONCLUSIONS

Delivery of ifosfamide daily for 3 days is feasible and safe at recommended dosages of 2133 mg/m2/d for children with prior exposure to platinum and 3000 mg/m2/d for those without prior exposure.

The Caenorhabditis elegans unc-31 gene affects multiple nervous system-controlled functions

We have devised a method for selecting Caenorhabditis elegans mutants that execute feeding motions in the absence of food. One mutation isolated in this way is an allele of the gene unc-31, first discovered by S. Brenner in 1974, because of its effects on locomotion. We find that strong unc-31 mutations cause defects in four functions controlled by the nervous system. Mutant worms are lethargic, feed constitutively, are defective in egg-laying and produce dauer larvae that fail to recover. We discuss two extreme models to explain this pleiotropy: either unc-31 affects one or a few neurons that coordinately control several different functions, or it affects many neurons that independently control different functions.

The genetics of feeding in Caenorhabditis elegans

The pharynx of Caenorhabditis elegans is a nearly self-contained neuromuscular organ responsible for feeding. To identify genes involved in the development or function of the excitable cells of the pharynx, I screened for worms with visible defects in pharyngeal feeding behavior. Fifty-two mutations identified 35 genes, at least 22 previously unknown. The genes broke down into three broad classes: 2 pha genes, mutations in which caused defects in the shape of the pharynx, 7 phm genes, mutations in which caused defects in the contractile structures of the pharyngeal muscle, and 26 eat genes, mutants in which had abnormal pharyngeal muscle motions, but had normally shaped and normally birefringent pharynxes capable of vigorous contraction. Although the Eat phenotypes were diverse, most resembled those caused by defects in the pharyngeal nervous system. For some of the eat genes there is direct evidence from previous genetic mosaic and pharmacological studies that they do in fact affect nervous system. In eat-5 mutants the motions of the different parts of the pharynx were poorly synchronized. eat-6 and eat-12 mutants failed to relax their pharyngeal muscles properly. These pharyngeal motion defects are most easily explained as resulting from abnormal electrical excitability of the pharyngeal muscle membrane.

Motor neuron M3 controls pharyngeal muscle relaxation timing in Caenorhabditis elegans

1. Previous work has shown that 12 of the 14 types of neurons in the Caenorhabditis elegans pharyngeal nervous system are collectively but not individually necessary for the trapping and transport of bacteria. The aim of these experiments was to determine the functions of individual neuron types by laser-killing combinations of neurons and looking at the effects on behavior. 2. The motor neuron M3 and the sensory neuron I5 are important in trapping bacteria, as shown by two observations. First, when M3 and I5 are both killed, trapping is inefficient in the isthmus (the middle section of the pharynx). Second, M3 is sufficient in the absence of the other 11 neuron types for normal trapping in the corpus (anterior pharynx). 3. M3 and I5 influence the timing of pharyngeal muscle motions. When M3 is killed, pump duration (the interval from the beginning of pharyngeal contraction to the end of relaxation) increases from 170 to 196 ms. This increase is at least partially due to a slower relaxation. Thus, M3 speeds up relaxation. Pump duration decreases to 159 ms when I5 is killed. When I5 and M3 are both killed, pump durations are long (192 ms), just as when M3 alone is killed. These observations, together with previous electron microscopic work showing synapses from I5 to M3, suggest that I5 slows down relaxation by inhibiting M3. 4. To explain these results, I propose that M3 and I5 promote bacterial trapping by regulating the relative timing of muscle relaxation in different regions of the pharynx.

Phase I study of flavone acetic acid (NSC 347512, LM975) in patients with pediatric malignant solid tumors

To evaluate the anticancer agent flavone acetic acid (FAA), we conducted a Phase I trial involving 17 pediatric patients with various malignant solid tumors. Dosages investigated included 5,120 and 6,144 mg/m2 given as 3-hour intravenous infusions; and 10,000, 12,500, 15,000, and 17,500 mg/m2 delivered in a 24-hour constant infusion with alkalinization. Grade 2 or worse toxicity was minimal, with 2 patients having nausea/vomiting, 2 having diarrhea, 1 becoming hypertensive, 1 becoming hypotensive, and 2 having myalgia. Three patients who received a 17,500 mg/m2 dose had no toxicity. Disease was stabilized for a brief period in 2 patients--1 with brain stem glioma and 1 with astrocytoma. The FAA pharmacokinetics varied with an average (SD) terminal half-life of 27.9 hr (18.7), clearance of 2.04 L/hr/m2 (0.37), and steady-state volume of 19.9 L/m2 (10.6). This study was discontinued because FAA caused no significant toxicity or therapeutic responses at doses 2.5 gm/m2 greater than had been tolerated by adults.

Ifosfamide, Fanconi's syndrome, and rickets

Three of 218 children treated with ifosfamide plus the uroprotectant mesna, in single- or combination-agent protocols, have developed Fanconi's renal syndrome, all of whom were in a subgroup of 86 children who had also received cisplatin or carboplatin therapy. Patients receiving ifosfamide who have received prior cisplatin (or carboplatin) are at significantly higher risk of developing Fanconi's syndrome than are those who have received no prior nephrotoxic therapy (P = .04). The role of prior nephrotoxic therapy, including cisplatin and its derivatives, and the total dose of ifosfamide should be considered in the assessment of this rare but serious and apparently irreversible side effect.

Effects of starvation and neuroactive drugs on feeding in Caenorhabditis elegans

Caenorhabditis elegans concentrates its food, bacteria, by pharyngeal pumping. The rate of pumping is affected by the presence of bacteria. Using a new assay that allows measurement of pumping rate in a population of worms suspended in liquid by measuring their uptake of microscopic iron particles, we have confirmed and quantitated this effect. Furthermore, we demonstrated that starvation stimulates pumping. Worms that had been deprived of bacteria for more than 4 hours pumped in the absence of bacteria under conditions in which well-fed worms did not. Furthermore, starved worms responded to lower amounts of bacteria than did fed worms. The assay was also useful for measuring effects of drugs on pumping. Of about 30 chemicals screened, 5 had clear effects. The neurotransmitter serotonin and the serotonin uptake inhibitor imipramine stimulated pumping, while the serotonin antagonist gramine inhibits. Imipramine stimulation is greatly decreased in cat-1 and cat-4 mutants, which have low levels of serotonin. Muscimol, an agonist for the neurotransmitter GABA, and ivermectin, whose site of action may also be the GABA receptor, both inhibit pumping. Qualitative observations suggested a role for acetylcholine in the regulation of pumping.

Pharyngeal pumping continues after laser killing of the pharyngeal nervous system of C. elegans

Using a laser microbeam to kill specific subsets of the pharyngeal nervous system of C. elegans, we found that feeding was accomplished by two separately controlled muscle motions, isthmus peristalsis and pumping. The single neuron M4 was necessary and sufficient for isthmus peristalsis. The MC neurons were necessary for normal stimulation of pumping in response to food, but pumping continued and was functional in MC- worms. The remaining 12 neuron types were also unnecessary for functional pumping. No operation we did, including destruction of the entire pharyngeal nervous system, abolished pumping altogether. When we killed all pharyngeal neurons except M4, the worms were viable and fertile, although retarded and starved. Since feeding is one of the few known essential actions controlled by the nervous system, we suggest that most of the C. elegans nervous system is dispensable in hermaphrodites under laboratory conditions. This may explain the ease with which nervous system mutants are isolated and handled in C. elegans.

Phase I study of 4'-deoxydoxorubicin (esorubicin) in children with malignant solid tumors

4'-Deoxydoxorubicin was given to 15 patients with drug-resistant pediatric malignant solid tumors with the objectives of determining the maximum tolerated dosage and dose-limiting toxicity. Maximum tolerated dosage was 36 mg/m2 given IV once every 3 weeks. Dose limiting toxicity was myelosuppression, which was severe and prolonged. Therapeutic benefits were not observed for these patients.

A cell that dies during wild-type C. elegans development can function as a neuron in a ced-3 mutant

Mutations in the C. elegans gene ced-3 prevent almost all programmed cell deaths, so that in a ced-3 mutant there are many extra cells. We show that the pharyngeal neuron M4 is essential for feeding in wild-type worms, but in a ced-3 mutant, one of the extra cells, probably MSpaaaaap (the sister of M4), can sometimes take over M4's function. The function of MSpaaaaap, unlike that of M4, is variable and subnormal. One possible explanation is that its fate, being hidden by death and not subject to selection, has drifted randomly during evolution. We suggest that such cells may play roles in the evolution of cell lineage analogous to those played by pseudogenes in the evolution of genomes.

Phase I clinical and pharmacokinetic study of bisantrene in refractory pediatric solid tumors

Fourteen patients with pediatric malignant solid tumors, median age 15 years, received 22 courses of bisantrene in a Phase I study. Dosage escalations ranged from 10 to 120 mg/m2 daily for 5 consecutive days. Toxicity included myelosuppression and phlebitis. A sensitive (detection limit of 2 ng/ml) and specific HPLC method was developed to quantitate bisantrene in patient's plasma and urine. Peak plasma concentrations at the end of 60 minute infusions ranged from 568 ng/ml at 10 mg/m2 to 6800 ng/ml at the 100 mg/m2 dosage. The elimination half life (T 1/2 beta) averaged about 10 hours but increased to 20 hours in a patient with liver disease. Only 2.4 - 10% of the bisantrene dose was eliminated in the urine suggesting that the liver may be the major route of elimination for this antineoplastic anthracene derivative.

In situ transposon replacement and isolation of a spontaneous tandem genetic duplication

Using a specialized transducing P1 phage carrying an insertion of Tn5-132, an insertion of Tn5-wt in the chromosome of Myxococcus xanthus, which codes for resistance to kanamycin, can be replaced with one of Tn5-132, which codes for resistance to tetracycline. That Tn5-132 in the daughter is inserted at the same location in the chromosome as Tn5-wt was in the parent was shown by a variety of physical and genetic tests. Southern blot hybridizations of restriction digests of daughter and parent DNAs probed for sequences homologous to Tn5 show that the physical location is the same. When KmR was transduced from the parent to the TcR daughter by the generalized transducing myxophage Mx4 or Mx8, all the transductants were TcS. Likewise, when the daughter was used as donor, TcR transductants of its KmR parent were KmS. Flanking markers that were linked to KmR in the parent were linked to TcR in the daughter. Spontaneous tandem genetic duplications of portions of bacterial chromosomes can be trapped by transducing a selectable marker from a donor to a recipient that has a different selectable marker at the same genetic location and selecting transductants with both markers. Using Tc-replacement, this technique can be applied to any region of the chromosome. We used it to isolate a spontaneous tandem duplication of part of the M. xanthus chromosome. The duplication was characterized by Southern blot hybridizations probed for Tn5-homologous DNA. It was also shown to be unstable by quantitation of loss of drug resistance. Transduction of the novel joint led to reconstruction of the duplication in the recipient strain. All these tests gave results consistent with the proposed structure. The methods described here are applicable to any bacterium into which transposons can be introduced, and for which some means of genetic exchange is available.