Medical resource use and cost of different first-line treatments for metastatic colorectal cancer in Brazil

Abstract

Background: Metastatic colorectal cancer (mCRC) is one of the most common malignancies worldwide. The availability of new chemotherapeutic agents have modified the treatment of mCRC over the years creating the need to evaluate the financial impact of treatment. The aim of this study was to establish and quantify the financial resources needed during the first-line treatment of mCRC in Brazil.

Methods: The authors began by reaching expert consensus using a modified Delphi panel with oncologists working at public and private services in Brazil. Costs were calculated using official databases and the microcosting technique.

Results: The panel reached consensus on six regimens used in the first-line treatment of mCRC, as well as the resources involved in the administration of these regimens. All the regimens contain either fluorouracil (5-FU)/leucovorin or capecitabine, combined with either oxaliplatin or irinotecan. The analysis showed that, when compared with intravenous 5-FU/leucovorin, the cost of capecitabine was offset by administration costs.

Conclusion: The panel concluded that regimens containing capecitabine, especially capecitabine plus oxaliplatin (XELOX) are less expensive than those containing 5-FU/leucovorin. Given the comparable efficacy and good tolerability of the XELOX regimen, it may be an attractive choice for the first-line treatment of Brazilian patients with mCRC.

• capecitabine
• catheterisation
• central venous
• colorectal neoplasms
• Delphi technique
• pharmacoeconomics

Introduction

Colorectal cancer (CRC) is one of the most common malignancies in the world, accounting for approximately 945,000 new cases each year. It is also the second most frequent cancer in developed countries1. In Brazil, the estimated number of new cases of CRC in 2006 was 11,390 in men and 13,970 in women2. Although the disease is curable when detected at an early stage, approximately 25% of patients present with metastatic CRC (mCRC), and such patients have a 5-year relative survival rate of only 10%. This poor prognosis is associated with high direct, indirect and intangible costs for patients, payers and the society in general.

Advances in diagnostic and therapeutic techniques, and also the availability of several new active chemotherapeutic agents, have modified the treatment of mCRC over the last 15 years. Fluoropyrimidines, represented by fluorouracil (5-FU), remained the only agents used by medical oncologists in this setting for almost four decades. In an attempt to increase efficacy, 5-FU has been modulated and administered in various ways, with only modest improvements3,4. This paradigm has changed significantly in recent years, due in large part to the introduction of multiple drugs that have demonstrated activity in the setting of mCRC5. The currently available active agents include the fluoropyrimidines 5-FU and capecitabine6,7, the topoisomerase I inhibitor irinotecan8, and the alkylating platinum oxaliplatin9. In addition, biological therapies, such as the angiogenesis inhibitor bevacizumab10 and the epidermal growth factor receptor-targeting monoclonal antibodies cetuximab11 and panitumumab12, have also shown efficacy in treating advanced disease; however, they have been approved only recently in Brazil and thus were not considered in the analysis. The current standard treatment for mCRC includes the use of a chemotherapy doublet in the first line, followed by a single agent or combination in the second and subsequent lines; whenever possible, biological therapies are combined5.

Capecitabine is an oral fluoropyrimidine developed to overcome the difficulties associated with intravenous administration of 5-FU and to increase the efficacy and tolerability of 5-FU-based chemotherapy13. Thymidine phosphorylase is the final step in a three-step enzymatic process that converts capecitabine into 5-FU14. During the time when capecitabine was undergoing investigation in Phase III trials, data were emerging on the superior efficacy of doublet regimens of 5-FU/leucovorin (LV) in combination with oxaliplatin or irinotecan compared with 5-FU/LV alone13.

A number of Phase III trials comparing 5-FU/LV/oxaliplatin (FOLFOX) with capecitabine/oxaliplatin (XELOX) have been conducted. In the TREE 1 trial, the median overall survival was 19.2 months for patients treated with FOLFOX, 17.9 months with bolus 5-FU/oxaliplatin and 17.2 months with XELOX15. The Phase III NO16966 trial showed that the median progression-free survival achieved with XELOX with or without bevacizumab was not inferior to that with FOLFOX-4 (8 vs. 8.5 months)16. That study also showed equivalence between XELOX and FOLFOX-4 in terms of overall survival (19.6 versus 19.8 months, respectively)17. Moreover, the ML16987 Phase III trial disclosed that XELOX is non-inferior to FOLFOX-6, since it has a higher response rate with a comparable safety profile18. The FOLFOX-4 regimen has well-known efficacy and toxicity profiles, but the FOLFOX-6 regimen is one of the most used due to its convenience in administration. XELOX is an alternative to FOLFOX-4 and -6 as first-line therapy in mCRC and offers the option of oral fluoropyrimidine administration17,18. Finally, the combination of capecitabine plus irinotecan (XELIRI) has demonstrated significant efficacy but has overlapping gastrointestinal toxicity14.

Clinicians are now faced with a number of different questions, including what is the most appropriate way to administer these various agents and what is the optimal combination of the many active therapies. The answers to these questions are central to improving treatment outcomes for metastatic disease, since the primary goal in treating these patients is prolonging life while minimising toxicity. The authors believe that besides efficacy and toxicity, economic issues assume utmost importance when choosing a chemotherapy regimen, since drug costs for treating mCRC may now vary up to 500-fold19. Therefore, they attempted to evaluate the treatment costs for commonly available chemotherapy doublets used in the first-line treatment of patients with mCRC.

Methods

The Delphi panel

A modified Delphi technique was used, in order to reach consensus regarding the current approach to mCRC in Brazil20,21. The goal was to capture the reality of the Brazilian healthcare system. Consequently, oncologists who worked in the public and/or the private healthcare service, with large experience in CRC and who follow a substantial number of patients were targeted. First, a sample with more than 30 expert medical oncologists was randomly assigned for the first phase of the study. The experts were addressed from the associated oncologists from the Brazilian Society for Clinical Oncology (SBOC). The experts answered a structured questionnaire about clinical patterns for mCRC treatment, which was developed in cooperation with a market research team with recognised expertise on this type of survey. Second, the results were gathered, analysed and submitted to a validation panel. Finally, a validation panel was conducted.

The validation panel was composed of six medical oncologists with recognised expertise among their peers in Brazil. These medical oncologists, who represented public and private services, saw at least 15–30 patients with mCRC per week, had worked in this setting for 5–19 years, and had consistent knowledge about clinical research in mCRC, reflected by meeting presentations, scientific papers, and/or membership in groups with special interest in this area. The panel was conducted in March 2006.

The panel aimed at establishing and quantifying the financial resources needed during the first-line treatment of patients with mCRC, including the disease itself and of the major complications derived from the use of a central venous device (CVD), which is typically required in patients using continuous infusions of 5-FU. The authors attempted to explore and to reach consensus regarding procedures, laboratory tests, drug use, and number and type of different health professionals involved in treatment. A contract research organisation was selected to act as moderator/facilitator for the questionnaire and panel meeting exercise. A two-round modified Delphi approach was used. First, a set of tables displaying mCRC regimens was provided to each panel member, along with the first-round questionnaire. These regimens, shown in Table 1, were selected initially from a review of the literature^{9, 22–24}. Since one of the panelists suggested the addition of one new regimen that was considered valuable (FLOX, bolus 5-FU, LV and oxaliplatin), this regimen was disseminated to the other panelists25. Questionnaires were based on the collected literature, current practice patterns and clinical experience with mCRC. In the second round, the oncologists met for a consensus meeting, which was conducted by a facilitator with previous experience in consensus-building strategies.

Table 1.  Regimens currently used in the first-line treatment of metastatic colorectal cancer.

Drugs	Regimen
	FOLFOX 4^9	FOLFOX 6^22	XELOX^23	XELIRI^24	FOLFIRI^22	FLOX^25
5-FU/LV	LV 200 mg/m^2 followed by 5-FU 400 mg/m^2 iv bolus on day 1; 5-FU 600 mg/m^2 iv CI for 22 hours on days 1 and 2	LV 400 mg/m^2 followed by 5-FU 400 mg/m^2 iv bolus on day 1; 5-FU 2.4–3.0 g/m^2 iv CI for 46 hours			LV 200 mg/m^2 followed by 5-FU 400 mg/m^2 iv bolus on day 1; 5-FU, 600 mg/m^2 iv CI for 22 hours on days 1 and 2	LV 500 mg/m^2 and 5-FU 500 mg/m^2 bolus weekly on weeks 1–6
Capecitabine			1 g/m^2 po twice daily for 14 days	1 g/m^2 po twice daily for 14 days
Oxaliplatin	85 mg/m^2 iv on day 1	100 mg/m^2 iv on day 1	130 mg/m^2 iv on day 1			85 mg/m^2 iv on weeks 1, 3 and 5
Irinotecan				240 mg/m^2 iv on day 1	180 mg/m^2 iv on day 1
Course duration	Every 2 weeks	Every 2 weeks	Every 3 weeks	Every 3 weeks	Every 2 weeks	Every 8 weeks

5-FU, fluorouracil; CI, continuous infusion; LV, leucovorin; iv, intravenous; po, by month.

Cost calculation and data analysis

Public databases were used to obtain the unit costs of each resource used in the first-line treatment of patients with mCRC. Costs of medications used were taken from Brasíndice, which publishes the official price list in Brazil26; the National Fees List of the Brazilian Medical Association27 was the basis for the costs of medical fees; costs of materials were taken from the Simpro Magazine28; and hospital rates were taken from the survey conducted by PROAHSA29. Costs related to the use of equipment and chemotherapy preparation were provided by the physicians in the Delphi Panel as there is no official source for these.

To determine the cost of a certain procedure or complication the microcosting technique30 was applied: every resource (physician time, medicines, materials, equipments, hospital structure etc.) was multiplied by its unit cost and then summed, resulting in the total value per therapy. Costs were presented in Brazilian reais (US$1 = R$2.12374, from 3rd January 2006).

Results

Chemotherapy regimens

Even though the frequency of use of the regimens shown in Table 1 varied according to the service, there was a greater trend towards the use of either FOLFOX-4 or FOLFOX-6 among the panel members. These are biweekly regimens, and their efficacy had already been demonstrated in clinical trials by the time the panel was held. Health insurance companies in Brazil typically cover both of these regimens. The FOLFIRI regimen, although used less often by the panel members, has the same characteristics as FOLFOX-4 in terms of treatment setting and coverage. Most patients receiving the FOLFOX-4, FOLFOX-6 and FOLFIRI regimens could theoretically be treated on an outpatient basis if portable infusion pumps were available. However, this is not always the case in Brazil, and a large proportion of such patients will need hospital admission when using regimens (assumed as 80%). This assumption has rather slight impact on total cost for these regimens. In the XELOX and XELIRI regimens chemotherapy is administered every 3 weeks, saving precious time for the patient. They also have the advantage of substituting the continuous infusion of 5-FU by the oral administration of capecitabine, thus avoiding the portable infusion pump issue; however, many health insurance companies in Brazil do not pay for oral drugs. Finally, the FLOX regimen requires no admission, since it employs only bolus administration of 5-FU, and is typically covered by health insurance companies. Despite these characteristics, the consensus concluded that this is the most expensive of the six regimens.

Central venous devices

One important point investigated in this study pertains to the use and complications of Port-a-Cath CVD. The panel members considered that the regimens FOLFOX-4, FOLFOX-6 and FOLFIRI require a CVD for their use in 100% of patients. With the XELOX, XELIRI and FLOX regimens, however, the panelists agreed that only between 20 and 30% of patients would need a CVD. Despite the use of a CVD in some patients treated with the XELOX and XELIRI regimens, the panelists concluded that catheter manipulation in these cases would be infrequent, with a resulting low rate of complications. It was considered that patients with a CVD would use it for a minimum of 12 months.

Ancillary drugs

According to the panel members, the antiemetic regimens used before intravenous administration of chemotherapeutic agents was the same for all six regimens. Such regimens typically include dexamethasone and ondansetron. On the other hand, the panel estimated that oral use of antiemetics at home would be required in only approximately 40% of the patients. In such cases, the choice of antiemetic depends on physician/service preference and on cost considerations, since these drugs are not covered by health insurance companies in Brazil.

For the regimens containing irinotecan (FOLFIRI and XELIRI), there was no consensus on the use of subcutaneous or intravenous atropin for diarrhoea prophylaxis, with rates varying among panel members from 30 to 50% in both regimens. On the other hand, there was consensus on the use of drugs for anaemia prophylaxis, with panel members prescribing erythropoietin (40.0000 IU/week) in 10% of the patients with all six regimens. Ferrous sulfate was used in 20% of the patients, and folic acid in only 10%, regardless of the chemotherapy regimen chosen. Likewise, there was consensus about the prophylaxis of neutropenia. For all regimens, the rate of use of filgrastim (300 mg/day for 5–10 days) was 5%, generally after 5 months from the beginning of treatment.

The intravenous use of magnesium sulfate and calcium gluconate was considered necessary in 80% of patients receiving three oxaliplatin-containing regimens (FOLFOX-4, FOLFOX-6 and XELOX) regimens. For the FLOX regimen, the use of these supplements was judged by the panel to be necessary in only 20% of the patients.

Diagnostic procedures

The diagnostic procedures required during chemotherapy were considered the same by panel members for all six regimens. Such procedures include blood tests for complete blood counts, renal function and hepatic enzymes. Imaging studies included a chest radiograph and computed tomography (CT) of the chest and abdomen. Imaging with CT was repeated every 2 months. Bone scan was not considered a routine examination, and was indicated only for patients with bone pain.

Costs

Table 2 shows the professional resources consumed by each of the six first-line treatment regimens. Table 3 lists all hospitalisations and materials used in a typical chemotherapy administration, including the use of gloves, syringes, saline/dextrose solutions, intravenous filters and lines, adhesive tapes and gauze. The total drug cost per regimen is shown in Table 4. Table 5 depicts the estimated cost associated with the most common complications resulting from a CVD (FOLFOX-4, FOLFOX-6 and FOLFIRI), and the estimated total cost for each of the six regimens is shown in Table 6, assuming a 6-month timeframe for first-line therapy. Costs of adverse events were not included in the analysis as negligible differences were addressed. Results were most susceptible to drug costs variation, both chemotherapic and non-chemotherapic drugs, which were set at public list prices. Other variables, such as procedures and body surface area were tested within a multi-way sensitivity analysis, depicted in Figure 1. Tested variables were complication costs, body surface area, acquisition costs and hospital costs, which were varied within ranges identified by the Delphi Panel and a 1.5–1.9m2 interval for body surface area.

Table 2.  Professional costs for each first-line treatment regimen used in metastatic colorectal cancer in Brazil for 6 months.

Cost component	Regimen
	FOLFOX-4	FOLFOX-6	XELOX	XELIRI	FOLFIRI	FLOX
Professional fees
Pharmacist fee	R$30.00	R$30.00	R$15.00	R$30.00	R$30.00	R$30.00
Hours	1 hour	1 hour	30 minutes	1 hour	1 hour	1 hour
R$/hour	R$30.00	R$30.00	R$30.00	R$30.00	R$30.00	R$30.00
Nursing fee	R$150.00	R$180.00	R$45.00	R$45.00	R$180.00	R$180.00
Hours	4–6	4–8	1–2	1–2	4–8	6
R$/hour	R$30.00	R$30.00	R$30.00	R$30.00	R$30.00	R$30.00
Medical oncologist fee	R$80.00	R$80.00	R$40.00	R$40.00	R$80.00	R$160.00
Visits/month	2	2	1	1	2	4
R$/visit	R$40.00	R$40.00	R$40.00	R$40.00	R$40.00	R$40.00
Vascular surgeon fee for CVC implantation (R$)	R$1,250.00	R$1,250.00			R$1,250.00

CVC, central venous catheter.

Table 3.  Hospitalisation and material costs for each treatment and for 6 months.

Cost component	Regimen
	FOLFOX-4	FOLFOX-6	XELOX	XELIRI	FOLFIRI	FLOX
Hospitalisation costs
Chemotherapy package cost/cycle	R$200.00	R$200.00	R$200.00	R$200.00	R$200.00	R$200.00
In patient admission total cost	R$1,050.00	R$1,050.00			R$1,050.00
Outpatient cost	R$42.00	R$38.00	R$42.00	R$42.00	R$42.00	R$42.00
Material costs
Laboratory procedures/cycle	R$459.44	R$459.44	R$459.44	R$459.44	R$459.44	R$459.44
CVC and Cytocan needle	R$1,196.12	R$1,196.12			R$1,196.12
Central venous kind (Porth-a-Cat)	R$256.00	R$256.00	R$64.00	R$64.00	R$256.00	R$64.00
Others (needles, syringes, normal saline, lines, gauze, adhesive tapes, gloves)	R$156.78	R$156.78	R$108.33	R$108.33	R$156.78	R$183.64
Infusion pump – reusable	R$952.00	R$952.00			R$952.00
Infusor	R$300.00	R$300.00	R$300.00	R$300.00	R$300.00	R$300.00

CVC, central venous catheter.

Table 4.  Drug costs for each first-line treatment regimen for 6 months.

Cost component	Regimen
	FOLFOX-4	FOLFOX-6	XELOX	XELIRI	FOLFIRI	FLOX
Non-chemotheraputic drugs	R$2,157.28	R$2,157.28	R$2,099.07	R$2,097.34	R$2,157.57	R$2,096.50
Antiemetics	R$186.88	R$186.88	R$126.65	R$126.65	R$186.88	R$126.65
Diarrhoea prophylaxis			R$2.02	R$2.25	R$2.25
Anaemia prophylaxis	R$1,654.63	R$1,654.63	R$1,654.63	R$1,654.63	R$1,654.63	R$1,654.63
Neutropenia prophylaxis	R$167.77	R$167.77	R$167.77	R$165.81	R$165.81	R$167.22
Disposable serum bag	R$148.00	R$148.00	R$148.00	R$148.00	R$148.00	R$148.00
Chemotherapeutic drugs/m^2/cycle	R$2,604.19	R$3,912.87	R$3,561.59	R$2,972.90	R$2,867.93	R$11,457.20

Table 5.  Estimated costs of the main complications resulting from the use of a central venous catheter and estimated rate of occurrence.

Complications	Obstruction	Infection	Thrombosis	Pulmonary embolism
Costs	R$902.47	R$3,407.54	R$4,038.30	R$24,675.44
Rate of occurrence	5%	2%	5%	1%
Expected cost of complication	R$45.12	R$68.15	R$201.92	R$246.75

Table 6.  Estimated total cost for 6 months of treatment with each first-line regimen in Brazil.

Treatment	FOLFOX-4	FOLFOX-6	XELOX	XELIRI	FOLFIRI	FLOX
Total costs*	R$66.394,16	R$95.341,99	R$55.846,93	R$47.186,84	R$72.253,10	R$67.016,61

Including costs of main complications

* Assuming a 1.7 m² patient for 6 months.

Figure 1.  Multi-way sensitivity analysis for six treatments comparison for 6 months.

Discussion

The results of the panel were in line with local guidelines from SBOC31, suggesting a good accuracy of the methodology used, especially considering that SBOC guidelines were published in 2007. The main difference between the panel and society guidelines was the under use in clinical practice of biological therapies, and as stated before, the access to these medications is limited for the vast majority of patients in Brazil.

Hence, the financial costs involved in the management of mCRC in Brazil were assessed, comparing six identified regimens used for the first line in Brazil. The cost analysis suggests that, despite the higher cost of oral capecitabine, this is mostly compensated for by other costs incurred with regimens containing continuous infusions of 5-FU. Such costs are due in part to hospitalisations for drug administration, a particular feature of the Brazilian health insurance system. In addition, the regimens containing continuous infusions of 5-FU may require portable infusion pumps. Nevertheless, the most significant cost difference suggested by the panel is the high cost resulting from the use of a CVD. In addition to the costs of insertion and the catheter itself, a major point to consider involves the costs related to complications inherent in the use of a CVD in this clinical setting.

Advanced malignancies such as mCRC have a significant impact on health-related quality of life. Such impact is of particular importance in light of the fact that prognosis is generally poor and gains from chemotherapy are modest. The management of patients with mCRC is associated with significant costs to patients, insurers, the government and society in general. Major cost determinants in this setting include diagnostic procedures, monitoring/surveillance, hospitalisation, surgery, radiation therapy, chemotherapy and supportive care. The cost of treating mCRC has been estimated to be 2–3 fold higher than for patients with localised disease. Drug costs constitute a relatively small proportion of the overall cost of cancer care: the costs of hospital stays, outpatient care and other medical encounters generally accounted for >75% of the imputed costs in a retrospective analysis of resource use in patients with mCRC in five European countries32.

With the increasing number of new treatments available, there is an urgent need to determine the most cost-effective options in mCRC33. Several pharmacoeconomic studies have been conducted to determine if the comparative clinical efficacy and superior tolerability of capecitabine over 5-FU/LV6,7, as well as the improved convenience and patient preference regarding oral administration34, translate into cost savings in the treatment of mCRC. Phase II clinical trials and interim results of one Phase III study showed that combination therapy with capecitabine and oxaliplatin (XELOX) is effective and well tolerated, thus representing an acceptable alternative to 5-FU/LV in the first-line treatment of mCRC16. In addition, some economic analyses have indicated that the XELOX regimen will result in cost offsets compared with the FOLFOX regimen, primarily as a result of the higher incidence of treatment-related adverse events associated with the use of 5-FU/LV35,36.

More recently, the use of medical resources was analysed in the context of the NO16966 trial37,38. The use of CVD was higher in the group of patients treated with FOLFOX-4, in comparison with those treated with XELOX. The number of patients with at least one hospitalisation was similar with FOLFOX-4 and XELOX, while the use of XELOX led to a slightly higher number of admissions and an increased length of hospitalisation (half a day/patient higher). The two regimens were similar in terms of the amount of resource use needed to treat adverse events. The authors concluded that substituting oral capecitabine for continuous 5-FU reduces the number and duration of infusion visits and patient time costs, thus providing a more convenient regimen for them37. In a companion report, the authors have shown that total direct medical costs were similar for FOLFOX-4 and XELOX (US$45,800 vs. US$44,500). However, XELOX had higher drug costs, while FOLFOX-4 had higher drug administration costs38. On the other hand, the overall impact on medical resource use will vary depending on local patterns of care, especially with regard to the use of portable pumps and alternative FOLFOX regimens. Indeed, a cost-minimisation analysis conducted in France suggests that XELOX decreases hospital resource consumption, in comparison with FOLFOX-6, with approximately 30% lower number of total hospitalisations and a projected cost saving of approximately €4,88918.

Analyses such as the one presented herein are subject to the premises adopted and to regional variations in patterns of care and costs of healthcare resources. Nevertheless, the panel composed of 30 members is thought to represent the view of Brazilian medical oncologists with recognised expertise in the management of mCRC. According to this panel, the XELOX and XELIRI regimens are less expensive than those containing intravenous infusions of 5-FU. This result, coupled with the confirmation of the comparable clinical efficacy between capecitabine- and 5-FU-containing regimens, suggests that the XELOX regimen in particular may become an attractive first-line choice for the treatment of Brazilian patients with mCRC. Further research with different evaluation methods, such as prospective data collection alongside a clinical trial, would be valuable for confirming results found in this medical resource use study, therefore improving database available in the country.

It is important to reinforce that, as there is an evident lack of database material in developing countries such as Brazil, performing cost studies, such as the one presented, is essential for aiding decision making where scarcity of resources is somewhat evident.

Acknowledgements

Declaration of interest: This study was funded by F. Hoffmann-La Roche Ltd. ES, MS and AC have all received financial sponsorship by Roche Pharmaceuticals. RC and RAR have no financial relationships to disclose.