This review is not intended as a complete study of the nephrotoxicity of chemotherapy agents used in the treatment of cancer. The number of these drugs that are cytotoxic has considerably increased in the last few years and our information is incomplete for many of them. We therefore reviewed the observations reported in the literature. Cis-platinum, streptozotocin, methotrexate at high doses, mithramycin and mitomycin are highly nephrotoxic. Other drugs, such as nitrosoureas, celiptium are less nephrotoxic while some appear to rarely induce nephrotoxicity. Anticancer drug nephrotoxicity is characterized by its particular insidiousness, its time of occurrence and its evolution. Since no clinical manifestations accompany the lesions, nephrotoxicity must be sought routinely. It can occur early or late, may be constant as of the first course or appear only after a certain cumulative dose and even occasionally after such a long interval that its cause may appear to be in doubt. The severity of this nephrotoxicity ranges from the usual first minor urinary anomalies to terminal renal failure. The pathophysiogenic mechanisms of the nephrotoxicity remain in most cases obscure. The mode of penetration into the cells is not known. There are fewer data on the interaction between the toxic agent and the cellular metabolism. In most cases, the drug itself in unchanged form does not seem to be the causative agent, which appears rather to be its metabolite. These metabolites are not always identified. Thus nephrotoxicity of antitumoral agents has not been given sufficient attention. Only better knowledge of their action within the kidney will eventually lead to progress in preventing their harmful side effects.

Ovarian function was investigated in 17 patients aged 13 5/12 to 30 years who had received various types of combined chemotherapy without any irradiation. Ovarian insufficiency was found in 6 cases with amenorrhea (n = 5) or irregular menstruations (n = 1). There is a high risk of sterility in these cases although as described in one case, a normal pregnancy occurred in spite of evidence of ovarian failure. Cyclophosphamide seemed to be less harmful when given before puberty. Great variations in individual susceptibility for relatively low doses were observed with this drug. The combination with other drugs in some protocols might play a role in these cases. At variance with results reported in adults, the MOPP chemotherapy used in children with Hodgkin's disease did not induce ovarian dysfunction.

Therapeutic advances for certain cancers raise now the problem of long-term toxicity of the chemotherapy, especially on the gonads. The risk of involvement of the gonadic function in women depends on the type of medication used, the dose and duration of the treatment and mostly the age of the patient. After the age of 35, amenorrhea is almost constant and often irreversible. The younger the woman, the better the prognosis. Finally, it seems that post-chemotherapy pregnancies are normal, and so are the children born. However, the carcinogenic risks of the descendants and the risks of mutation of other generations are not well known at the present time.

One hundred and thirty-nine chambers for injection or infusion of cytotoxic drugs were implanted in 9 French anticancer centres, because of inadequate peripheral veins. A strict surgical procedure and a good training of the nursing staff are necessary to avoid complications (27.7%, including 7% obstructions). Complications associated with the material are exceptional. The system was very well tolerated by more than 90% of the patients. Totally implantable devices for intravenous chemotherapy constitute a major improvement over the previously used systems.

Twenty eight patients with isolated liver metastases of colorectal origin have been treated with discontinuous intra-arterial chemotherapy. This treatment was performed weekly with a surgically implanted subcutaneous access chamber. In all the patients the metastases were non-resectable and involved less than 75% of the liver. The first six patients were investigated in a phase I study which demonstrated that the method was well tolerated and provided a normal life at a lower price than the totally implanted pump. A subsequent phase II study included 22 patients. After implantation, they received an 8-hour infusion of 5 Fu 1 g/m2 each day for 8 consecutive days and after discharge, a weekly 8-hour infusion of 5 Fu 1 g/m2 and mitomycin C 1.5 mg/m2 as out-patients. Twenty-one patients have now been followed up for more than 3 months and are assessable in terms of response: we observed 4/21 complete responses, 6/21 partial responses, 3/21 minor responses, 7/21 stable disease and 1/21 progression. The objective response rate was 48%. The complication rate was low (2 leukopenias and 2 duodenal ulcerations) and comfort excellent with normal life between courses. In conclusion, with discontinuous intra-arterial chemotherapy we obtained the same response rate as with the totally implanted pump, with good tolerance and quality of life and perhaps a lower rate of complication. The two methods should now be compared in a randomized trial.

Determination of intramyocardial lactate dehydrogenase (LDH) and its isoenzymes (LDH 1 to 5) has been conducted on biopsy material obtained from the right ventricle in 18 patients who had received the theoretical maximal dose of anthracyclines. Total LDH activity, expressed in mU/mg of myocardial protein is higher (852.61 +/- 87.98) than in subjects with good left ventricular function (334 +/- 208), p less than 0.02. The activity of LDH 1 is decreased in the treated group, whereas that of LDH 3,4 and 5 is raised: LDH 1 = 41.95 +/- 7.25 per cent instead of 47.76 +/- 7.93 per cent (p less than 0.03); LDH 3 = 15.34 +/- 5.09 per cent instead of 9.18 +/- 4.56 per cent (p less than 0.05). The H/M ratio (H = fractions of heart isoenzymes; M = fractions of muscle isoenzymes of LDH) is decreased in the treated group: H/M = 4.06 +/- 1.13 instead of 5.30 +/- 1.80 (p less than 0.01). When the histological lesions are minimal (stage 1 or 1.5 of Billingham), the H/M ratio is not altered: 4.92 +/- 1.02. When the lesions attain or exceed stage 2, the H/M ratio is significantly decreased: 3.74 +/- 0.82. These results, which indicate an increase in anaerobic metabolism, add quantitative information to the histological study.