Dose banding was the topic of discussion at a stimulating satellite symposium sponsored by Sun Pharma, which featured a distinguished faculty from the UK, France and Germany, at the recent EAHP conference in Vienna
Dose banding was the topic of discussion at a stimulating satellite symposium sponsored by Sun Pharma, which featured a distinguished faculty from the UK, France and Germany, at the recent EAHP conference in Vienna
Rob Duncombe MSc BPharm(Hons) MRPharmS
The Christie NHS Foundation Trust,
Manchester, UK
Email: [email protected]
Support for the development of this report was provided by Sun Pharma
The concept of dose banding of chemotherapy has been known for many years, but it has recently returned to prominence as the benefits of dose banding are re-evaluated (and as commercially available products start to emerge).
The uptake of dose banding across Europe has been variable, and even in the same country, very different levels of uptake are observed.
Improving the patient experience
In the opening presentation, Professor Graham Sewell (University of Plymouth, UK) detailed the approach that he had taken to dose banding almost 20 years ago. This was initially driven by a need to improve patient experience and reduce waiting times; however, it progressed to being a mechanism by which chemotherapy could be supplied using a relatively small number of dosage units, and supported by strong scientific evidence.
Dose banding is not rounding of doses or capping of doses – it is a means of preparing doses of chemotherapy within set dosage bands. This means the majority of doses can be supplied by using a limited number of doses. Professor Sewell demonstrated the principles of dose banding with reference to the preparation of 5-fluorouracil (5-FU).
Example
Dose band: 875mg to 925mg.
Banded dose = 900mg.
Dose band: 925mg to 975mg.
Banded dose = 950mg.
Each dose band is 50mg higher than the previous dose. The maximum variance, for any dose is no more than 25mg from the prepared dose, resulting in a maximum variance from the dose prescribed of no more than ±5%.
A similar principle can be applied to almost all chemotherapeutic agents. Depending on the dosing regimen, a larger spread of dose-banded products may be required, but generally no dose is at a greater variance to that prescribed of ±5%.
Dose banding is not something that can be undertaken unilaterally within pharmacy but needs the concordance of all professionals involved in the prescribing and administration of chemotherapy. The science behind dose banding is quite compelling but does require the skill set and professional knowledge of the pharmacist to explain the logic and reasoning behind it to sometimes quite sceptical clinicians (doctors, nurses and fellow pharmacists).
Advantages
The advantages of dose banding are relatively straightforward:
- It allows pharmacy departments to manage increasing workloads
- It supports reductions in outpatient waiting times
- It reduces the risks of compounding errors
- It reduces drug wastage
- It enables a better use of resources
- Recognises that the current method for the dosing of chemotherapy is supported by limited scientific rationale
All hospitals are facing increasing demands on their aseptic units capabilities, aligned with some restrictions on the staffing resource available. Dose banding should permit there to be a greater focus on those activities which have to be undertaken in a local aseptic facility. (For example, complex parenteral nutrition, preparation of drugs with limited shelf lives and preparation of clinical trials).
The only cytotoxic drug that has a real scientific basis behind its dosing is carboplatin – where the pharmacological measure of the area under the curve (AUC) is used to support dosing.
The conventional means of dosing chemotherapy either on weight or surface area, has limited correlation to pharmacokinetic measures of a drug’s effectiveness, in terms of how a patient clears a drug, the achieved plasma levels or levels of the drug that actually are delivered to the tumour (with multiple other factors affecting this). Obviously this method of dosing does work and does deliver doses of chemotherapy to patients which successfully treat their cancers.
There are multiple sources of error in the delivery of a chemotherapy dose. (Delivery here is used in its broadest sense – as in delivery of a package of care or a treatment):
- Measurements of height and weight
- Errors in dose calculation
- Small variances in the manufactured concentration of the drug. (that is, Gemcitabine 100mg/ml will have an acceptable tolerance in its concentration. Say 98mg/ml–102mg/ml)
- Compounding errors
- Errors in administration (spillage, priming, syringe accuracy)
- Patient-related factors (organ function, drug handling).
Oral chemotherapy, while potentially just as toxic as intravenously administered chemotherapy, is never dosed to within such tight limits. For example, capecitabine is only ever dosed to the nearest 150mg.
Dose banding should also deliver further benefits. These include:
- Batch preparation
- Automated compounding
- The opportunity to use licensed dose banded products
- Minimising drug wastage through vial sharing
Points to consider
To introduce dose banding successfully, one needs to be able to demonstrate that this method of dosing of chemotherapy is both safe and effective. Therefore, consideration is required of the following:
- Is it safe for patients? Do patients receive sufficient drug to ensure therapeutic levels are achieved?
- Is it cost effective? Are we able to clearly demonstrate reduced wastage and better use of resources?
- Is it pharmaceutically safe? Are the dosage units chemically stable, free of microbial contamination and physically stable?
- Are clinicians willing to move to this method of dosing for chemotherapy?
If the above can be proven, then dose banding can, and should, be adopted for the majority of cytotoxic drugs in common use.
Is it safe for patients?
The three speakers presented data demonstrating how dose banding was safe for patients, and was at least as effective in delivering chemotherapy to tumours as the more conventional methods of dosing.
The faculty used different examples to show how dosing on the basis of BSA was an imprecise science. Priv-Doz Dr Hug presented data from a study by Samm and colleagues,1 which showed the plasma concentrations of 5-FU in patients treated for colorectal cancer, using BSA as a method for dose calculation. The distribution of concentrations of dose received was widely spread; if dosing by BSA was delivering consistent plasma levels, the expectation would be that the vast majority of patients would be receiving a very similar therapeutic dose, and there would be a very tight correlation between the dose given and the plasma concentration.
Professor Sewell presented results of research that he and his team had conducted on 26 patients receiving FEC chemotherapy.2 This was a prospective open-label, cross-over study in which patients received one course of treatment with the 5-FU dosed on the basis of the calculated BSA, and one course dosed using dose banding methodologies. The 5-FU AUC (area under the curve) was used as a surrogate for clinical effectiveness and toxicity. The results of this study demonstrated that there was no significant difference in the AUC between the two methods used for dosing of 5-FU.
While much of the scientific evaluation of dose banding has been focussed on 5-FU, possibly due to ease of measurement, it would appear reasonable to extrapolate these results to other agents. The conclusion to be drawn is that dosing by BSA does not deliver consistent plasma levels, and certainly not within the 5% margins of dosing proposed by the champions of dose banding.
Is it cost effective?
For dose banding to be cost effective, it must result in reduced wastage and reduced costs of production. There is little point in dose banding a drug if through batch production, large numbers of doses are going to be wasted. Therefore, when choosing candidates for dose banding, there has to be sufficient use of the drug to justify this, and the choice of dose bands has to be such that the majority of doses used fall within the bands chosen.
The speakers described the processes that they had used in deciding which drugs to dose band and the bands chosen.
Identifying candidates
Dr Frédéric-Antoine Chastang (Institut Curie, France) shared his criteria for identifying candidates for dose banding at his institute in Paris. These are:
- Production rate of >250/year
- A maximum of five doses
- Drug stability
- Volume of production easy to manage
- Homogenous cohorts of patients
For cyclophosphamide, this resulted in certain bands being developed (see Table 1).
In terms of wastage, the evidence presented by Dr Chastang showed a significant reduction in dose wastage between dose-banded preparations and individualised prepared units.
Furthermore, the introduction of dose banding permitted significant expansion in the delivery of chemotherapy services without similar increases in staffing resource. Priv-Doz Dr Hug shared the increase in workload of his hospital from 2001 through to 2015 where use of chemotherapy had almost doubled. Without dose banding such an increase in activity would have been very difficult to manage.
Is it pharmaceutically safe?
Priv-Doz Dr Martin Hug (University Clinic Freiburg, Germany) presented some of the work that his team had conducted to gain assurances that the dosage units prepared through dose banding were safe.
His team had undertaken a number of studies to show microbiological stability over prolonged periods. This entailed the use of simulated manipulations substituting tryptone soya broth in place of the chemotherapeutic agent, and then observing and testing for microbiological growth. These tests showed that when filled under aseptic conditions, the integrity of the final containers was assured.
Dr Hug and his team then looked for physical changes in filled containers, in terms of colour, pH and presence/absence of particulate matter. These examinations showed that the filling of dose-banded units provided physically stable solutions.
Finally, Dr Hug and his team undertook a very detailed assessment of the chemical stability of gemcitabine. This examination showed that in solution gemcitabine was chemically stable for at least 12 weeks with less than 5% loss of drug over a 12-week period. However, through more detailed analysis, a number of impurities were identified.
It is not entirely clear where these impurities came from, or their significance, but they were present in low concentrations, and were possibly either degradation products or impurities that had leached from the final container of the gemcitabine.
Do clinicians accept dose banding?
Without clinician acceptance, dose banding will always remain an unfulfilled pharmacy-led initiative.
All three speakers found that there were high levels of clinician acceptance of dose banding and perhaps the only area of disagreement is over the percentage size gap of the bands, whether it be 5% or 10%.
Conclusions
The key points are:
- Dose banding is a logical way to dose chemotherapy
- Dose banding is a safe way to deliver chemotherapy
- Oncologists support dose banding
- Dose banding has the potential to improve efficiency and maximise the best use of a limited resource
- Dose banding can reduce wastage.
In summary, it does appear that there are very few reasons not to be dose banding chemotherapy. It is safe, effective, logical and clinically acceptable. Certainly dose banding is not an end to compounding but a means by which hospital pharmacy teams are able to introduce evidence based efficiencies which permit a reallocation of a very expensive resource to those areas where it is needed most.
References
- Saam J et al. Body surface area-based dosing of 5-fluoruracil results in extensive interindividual variability in 5-fluorouracil exposure in colorectal cancer patients on FOLFOX regimens. Clin Colorectal Cancer 2011;10(3):203–6.
- Moores MJ et al. Variability in the pharmacokinetics of cyclophosphamide, methotrexate and 5-fluorouracil in women receiving adjuvant treatment for breast cancer. Cancer Chemother Pharmacol 1994;33(6):472–6.
