Patients and methods

Blood samples from children with ALL receiving ‘maintenance’ treatment were sent from all over the UK for assays of 6-mercaptopurine metabolites in red blood cells. They were taken at any time when 6-mercaptopurine had been prescribed at full protocol dose (or more) for at least seven days, by which time the intracellular metabolites should have been easily detectable and at or approaching steady state levels. In the small number of children where the assay values were zero, clinical information, including demographic and social factors, and an assessment of compliance were sought.

Results

A total of 496 children was studied. The range of metabolite concentrations seen (6-thioguanine nucleotides) ranged from 0 to 1410 pmol/10⁸ red blood cells, with a median value of 301. Nine children (2%) were found to have completely undetectable metabolites on one or more occasions. A further four children had less than 50 pmol/10⁸ (and were thus at the limit of the assay sensitivity), 10 had values between 50 and 100 pmol/10⁸, and 21 between 100 and 150 pmol/10⁸. All patients within this range would give cause for concern about the adequacy of treatment.

Of the nine children with a zero concentration of metabolites, six were boys and three girls. They were aged from 5 to 18 years (median 11 years). All but one were white Europeans. Four were from social classes 4 and 5 and two from classes 1 and 2. Of the fathers whose occupations were known, four were drivers (lorry, van, or taxi), and the others included work in business, environmental health, and psychology. Four of the mothers worked outside the home in business, psychology, journalism, and cleaning. Five of the nine children had older siblings, six had younger siblings, and three had both. Five (55%) were adolescents compared with 83 (17%) of the remaining children (χ² = 6.52, p < 0.02).

At the time the samples were taken the children had supposedly been taking from 67 to 188 mg/m² of daily 6-mercaptopurine without reduction for four to 52 weeks. All but one had completed more than one year of treatment. Their ages and the doctor’s assessment of the likelihood and possible reason for non-compliance are shown in table 1. There was no apparent reason to suspect non-compliance in only one patient. Of the nine children who did not comply with treatment, seven continue in their first remission and two of those are still receiving treatment. The remaining children have been followed for two to four years after the completion of treatment.

Discussion

An assay of 6-mercaptopurine metabolites in red blood cells has provided a convenient way of monitoring the adequacy of ‘maintenance’ treatment given by mouth in childhood ALL; those children with lower concentrations are at a higher risk of relapse.1 2 One of many reasons for low concentrations is poor compliance and we have a growing suspicion that this may be a more important problem than is generally appreciated.3 More children than the nine described here in detail produced inappropriately low or very low assay values that may well have been the result of intermittent or partial compliance, but alternative explanations cloud the issue. The total absence of metabolites is hard to explain on any other basis, however, so we chose to concentrate on those nine children, imagining them to be representative of the greater population of delinquent compliers.

We wanted to see if they formed any socially or clinically identifiable group, but, apart from there being more adolescents than would be expected by chance, and also that most had been receiving treatment for over a year (both previously suggested risk factors for non-compliance4), they were hard to categorise. In a previous regional study we noted an excess of children with Asian family names in a group of 32 children with inappropriately low metabolite levels,5 but only one of the nine children in this study was Asian. As for outcome, only two of the nine children had relapsed at the time of the study, though the numbers are too small and follow up too short for any conclusions to be drawn from this, except that they must have taken some of their drug some of the time. It may be the duration rather than the degree of non-compliance that is important, with persistent partial defaulters being more at risk than the occasional total abstainer.

We do not know whether the children described here are truly representative of all those who do not reliably take their chemotherapy drugs. All we can say is that adolescents should be watched carefully, particularly in the second year of treatment. More importantly, we emphasise that no social class or ethnic group appears to be immune from the problem and that patients with cancer appear to be no different from other children in this respect.