SciELO - Scientific Electronic Library Online

vol.41 número1Prescripción inapropiada en personas mayores: hora de pasar a la acciónReducción de la fatiga por alertas en un sistema de prescripción electrónica asistida, mediante la metodología Lean Six Sigma índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados




Links relacionados

  • En proceso de indezaciónCitado por Google
  • No hay articulos similaresSimilares en SciELO
  • En proceso de indezaciónSimilares en Google


Farmacia Hospitalaria

versión On-line ISSN 2171-8695versión impresa ISSN 1130-6343

Farm Hosp. vol.41 no.1 Toledo ene./feb. 2017 


Cost-effectiveness of pemetrexed in combination with cisplatin as first line treatment for patients with advanced non-squamous non-small-cell lung cancer in Spain

Jonathan González García1  , Fernando Gutiérrez Nicolás1  , Gloria Julia Nazco Casariego1  , Cristina Valcárcel Nazco2  , José Norberto Batista López3  , Juana Oramas Rodríguez3 

1 Pharmacy Unit, Complejo Hospitalario Universitario de Canarias. Spain.

2 Evaluation Department, Canarian Health Service. Spain

3 Oncology Unit, Complejo Hospitalario Universitario de Canarias. Spain.



Lung cancer is the third most frequent neoplas-tic tumour in Spain, with around 27 000 new cases diagnosed per year; 80-95% of these are non-small-cell cancer (NSCLC), and the majority of cases are diagnosed in advanced stages of the disease, and for this reason it is one of the oncologic conditions with higher mortality rates (21.4% mean survival at 5 years).

The main treatment regimens used for first-line treatment of NSCLC are: cisplatin/pemetrexed (cis/pem), cisplatin/gemcita-bine/bevacizumab (cis/gem/bev), and carboplatin/paclitaxel/ bevacizumab (carb/pac/bev). The objective of this study was to evaluate the cost-effectiveness ratio of antineoplastic 1st line NSCLC treatment regimens, from the point of view of hospital management.


A cost-efficacy mathematical model was prepa-red, based on a decision tree. The efficacy variable was Pro-gression Free Survival, obtained from the PARAMOUNT, AVAIL and SAIL Phase III clinical trials. The study was conducted from the perspective of the hospital management, considering only the direct costs of drug acquisition. A deterministic sensitivity analysis was conducted to confirm the robustness of outco-mes.


The PFS obtained in clinical trials with cis/pem, cis/ gem/bev and carb/pac/bev was: 6.9, 6.7 and 6.2 months, res-pectively. Based on our model, the mean cost of treatment per patient for these regimens was: 19 942 €, 15 594 € and 36 095 €, respectively. The incremental cost-effectiveness ratio per month of additional PFS between cis/pem and cis/gem/bev was 19 303 €. Estimating a 30% reduction in acquisition costs for pemetrexed (Alimta®Eli Lilly Nederland B.V.), due to the forthcoming launch of generic medications, the cis/pem treat-ment would become the predominant alternative for 1st line treatment of NSCLC patients, by offering the best health re-sults at a lower cost.

KEYWORDS: Non-small-cell lung cancer; Cost-effectiveness; Economic evaluation; Cisplatin; Pemetrexed; Bevacizumab

Contribution to scientific literature

This article evaluates the efficiency of using cisplatin/pemetrexed, cisplatin/gemcitabine/bevacizumab or carboplatin/paclitaxel/bevacizumab as first line treatment for patients with non-squamous non-small-cell lung cancer, from the perspective of the hospital management. The interest of this article lies in the fact that, as far as we know, no pharmacoeconomic evaluation has been conducted so far in order to evaluate the efficiency of using the main three first line treatment regimens for this condition, taking also into account the potential economic impact of the imminent launch of generic pemetrexed. Given our current economic situation, with limited healthcare resources, this type of studies appear as an additional tool to be considered at the time of making decisions, by complementing the information about efficacy and safety extracted from clinical trials.


After colorectal cancer and prostate cancer, lung cancer is the most frequent tumour in Spain, with around 27,000 new cases /year (13% of the total number)1; 80-85% of these are non-small-cell lung cancer (NSCLC) cases2. The majority of patients are diagnosed in advanced stages of the disease3, and for this reason it is one of the oncologic conditions with higher mortality rates (21.4% mean survival at 5 years)4.

The standard treatments for NSCLC are platinum-based antineoplastic regimens5, and in recent years, new agents have been incorporated for metastatic disease, such as pemetrexed and bevacizumab. Some regimens based only on chemotherapy doublets have also been described for the induction stage, followed by maintenance with pemetrexed; but guidelines currently recommend the incorporation of pemetrexed or bevacizumab since the induction stage6. Both drugs have demonstrated an improvement in the life expectancy of these patients, when used in treatment regimens including one platinum-based induction stage with pemetrexed or bevacizumab, followed by maintenance with any of these two drugs. These regimen are mostly: cisplatin/pemetrexed (Pem/Cis)7, cisplatin/gemcitabine/bevacizumab (Gem/Cis/Bev)8 and carboplatin/paclitaxel/bevacizumab (Pac/Carb/Bev)9. The improvement in efficacy represented by incorporating these two new drugs has also entailed an increase in the cost of treatments9. And the fact that the efficacy outcomes shown by the three regimens are similar, together with the lack of clinical trials comparing them with one another, means that currently it is difficult to pick the most efficient regimen. Therefore, particularly in our current economic situation, with limited healthcare resources, it is necessary to conduct a cost-effectiveness analysis to be used as a tool for decision making at the time of optimizing those resources available.

Therefore, the objective of this study was to conduct an economic analysis to estimate the cost-effectiveness ratio of first line antineoplastic regimens for NSCLC treatment, from the perspective of the hospital management.

Material and methods

Pharmacoeconomic model

A mathematical model for cost-effectiveness was prepared, based on a decision tree with three branches, each of them representing the cis/pem, cis/gem/bev and carb/pac/bev regimens (Figure 1). The probability values for each transition were obtained from these Phase III clinical trials: PARAMOUNT7 (cis/pem); AVAIL8 (cis/gem/ bev) and SAIL9 (carb/pac/bev).

These treatment regimens are based on an induction stage followed by a maintenance stage on monotherapy which is continued until disease progression (Progression Free Survival - PFS), or unacceptable toxicity.

4 Cis (75 mg/m2)/Pem (500 mg/m2) courses, followed by pemetrexed (500 mg/m2) as maintenance every 3 weeks.

6 Gem (1250 mg/m2)/Cis (80 mg/m2)/Bev (7.5 mg/ kg) courses, followed by bevacizumab (7.5 mg/kg) as maintenance every 3 weeks.

6 Pac (200 mg/m2)/Carb (AUC = 6)/Bev (15 mg/kg) courses, followed by bevacizumab (15 mg/kg) as maintenance every 3 weeks.

A summary of the main characteristics of the PARAMOUNT, SAIL and AVAIL clinical trials appears in Table 1.

Table 1 Outcomes and characteristics of the PARAMOUNT, AVAIL and SAIL clinical trials6,7,8 

The economic model was validated by an expert panel of oncologists, with the objective to clarify any differences and/or contradictions described in scientific literature. The Microsoft Excel 2013® computer application was used for developing the model. Figure 1 shows a schematic diagram of the decision analysis.

Estimation of efficacy and costs

The efficacy variable used was Progression Free Survival (PFS) as described in the clinical trials for each treatment. Given that in the AVAIL study, those patients who survived the induction stage were randomized to receive bevacizumab as maintenance at a 7.5 mg/kg or 15 mg/ kg dose, and no statistically significant differences were found between both bevacizumab doses, the 7.5 mg/kg dosing was used for cost estimation.

The pharmacological cost was estimated according to the LSP (Laboratory Selling Price) of the medications studied, resulting from applying the profit margins established on the retail price-VAT of the Catalog by the General Council of Official Pharmaceutical Colleges, and with the relevant 7.5% deduction on the LSP established by Royal Decree 8/2010 and provided by the Ministry of Health, Social Services and Equality on October, 201210,11 (Table 2).

Table 2 Costs per unit used in the pharmacoeconomic model for 1st line NSCLC treatment (at October, 2015 euros) 

G.M.P. Generic Medicinal Product

* 7.5% reduction on the LSP established by Royal Decree 8/201014

For dose calculation, we used the anthropometric characteristics of a standard patient representative of Spanish oncology patients12: weight of 70.15 kg, body surface of 1.78 m2, and serum creatinine of 0.6 mg/dL. The administration of carboplatin in order to obtain the desired area under the curve (AUC) was determined by the Glomerular filtration rate (GFR). Calvert Formula was used in this study to calculate the carboplatin milligrams needed for each administration13. The total pharmacological cost for each oncological regimen was estimated by assuming maximum use of vials. All costs included in the analysis were stated in 2015 euros (€).

Time horizon, analysis perspective, and discount rate

The time horizon considered was the time until progression to first line of treatment, which coincided approximately with 1 year of modelling based on clinical trial outcomes. The study was conducted from the perspective of the hospital management, taking into account only the direct costs for drug acquisition. Given the limited survival of patients (none of the oncology regimens went beyond one year of treatment), it was not adequate to apply a discount rate correction(14.15).

Base case and Sensitivity Analysis

The mean values of costs and efficacy were considered for the base case in the study.

In order to confirm the robustness of the base scenario outcomes and the consistency of the estimations, various univariate deterministic sensitivity analyses were conducted, considering the following factors:

Improvement in efficacy (PFS). The PFS of patients who survive the induction stage is not specified in clinical trials; therefore, it was estimated with the following formula:

The model assumes that the PFS of patients with progression during the induction stage is equal to the duration of said stage. This assumption is based on the fact that, in clinical practice, imaging tests to confirm the status of the disease are usually conducted after completing the induction stage. Therefore, the estimated PFS for patients who presented disease progression during the induction stage was:

2.8 months for the cis/pem regimen (4 induction courses every 3 weeks).

4.2 months for the pac/carb/bev and cis/gem/bev regimens (6 induction courses every 3 weeks).

A sensitivity analysis was conducted, assuming a 20% improvement in the proportion of patients who survived the induction stage for each regimen (Table 3); by applying the previous formula in an overall PFS for each regimen, this would result in:

cis/pem: 7.69 months

cis/gem/bev: 7.18 months

pac/carb/bev: 6.58 months

Table 3 Incremental cost-effectiveness ratio and sensitivity analysis for the cost-effectiveness analysis (at October, 2015 €) of 1st line NSCLC treatment 

a Incremental cost-effectiveness ratio per additional month of Progression Free Survival

b An alternative is defined as dominated when there are options with higher efficacy and lower cost.

c An alternative is defined as dominant when it presents the highest efficacy at the lowest cost.

3. Modification of the acquisition costs for pemetrexed, assuming a 30% cost reduction with the launch of generic drugs.


Base Case Result

The proportion of patients who moved on to the maintenance stage during clinical trials was: 57% for the cis/ pem regimen, 42% for patients who received gem/cis/bev, and 53% for the carb/pac/bev regimen. Based on our model, the mean cost of treatment per patient for the gem/cis/bev, cis/pem, and carb/pac/bev treatment regimens would be 15,594.74€, 19,442.01€ and 36,095.17€ respectively. The car/pac/bev regimen is the dominated alternative, because it is the most costly treatment regimen with the worst health outcomes. The difference in costs between the cis/pem and cis/gem/bev approaches is 3,847.21€, with an incremental cost-effectiveness ratio (ICER) per additional PFS month of 19,303€ (Table 3).

Sensitivity analysis in alternative scenarios

There was a robust outcome in the sensitivity analysis for all scenarios, considering the current drug acquisition costs; with an ICER of 9,720€ per additional PFS month for the cis/pem regimen, when a 20% increase in efficacy is considered for all treatments. However, with a 30% reduction in the price of pemetrexed, the cis/pem alternative would become the dominant alternative, because it presents the best health outcomes at a lower cost. The best incremental cost-efficacy ratio for the cis/pem treatment occurred when a 30% discount was assumed for pemetrexed, as well as a 20% increase in efficacy (ICER = -3,974€/additional PFS month ).

The carb/pac/bev regimen was the dominated alternative in all scenarios studied (Table 3).


The outcomes of our model show that, from the perspective of hospital management, the carb/pac/bev regimen is the dominated alternative in first line treatment for NSCLC, because there are current regimens with higher efficacy and lower cost; therefore, in our opinion, this option would be relegated only to patients who cannot tolerate cisplatin-based treatments; for example, those patients with underlying renal conditions. At the current acquisition cost for the drugs studied, the cis/ gemc/bev treatment is cheaper than the cis/pem treatment, with an incremental cost of between 8,217 and 19,303€ per additional PFS month. Typically, a therapy will be considered cost-effective if its ICER is below the predefined threshold of acceptability17. If no explicit thresholds of acceptability are available, a therapy can be considered cost-effective if its ICER is within the range of other therapies which healthcare managers have been willing to pay for16. In a study published in 2013 about the efficiency of oncologic treatments in Spain, Oyagüez et al. found that the ICER values per additional PFS month were within the range of 5,979.87 to 46,716€17 for first line therapies; therefore, according to these studies, the conclusion could be that the cis/pem treatment is a cost-effective strategy.

Besides what has been previously stated, the incorporation to the market of generic pemetrexed (Alimta®Eli Lilly Nederland B.V, date of authorization: September, 2004) can lead to a significant change in our current situation. With a 30% reduction in the current acquisition costs for pemetrexed, the cis/pem treatment would become the dominant alternative (the best health outcomes at a lower cost) for the 1st line treatment of NSCLC patients. These outcomes should be taken into account at the time of making decisions in order to optimize the resources available.

The interpretation of the outcomes of this analysis must be understood as an approach to the efficiency of oncologic regimens, and in any case it should be considered alongside a series of limitations. The main limitation is that only the pharmacological cost of treatment regimens has been considered, though some therapies could be associated with additional costs for the health system, such as those derived of their administration, monitoring, or management of adverse events experienced by patients and caused by medication. The approach chosen for this analysis, considering the pharmacological cost only, intends to equalize the outcomes for the different regimens evaluated. Therefore, the outcomes of this pharmacoeconomic analysis should be confirmed through effectiveness records, comparing the use of healthcare resources for the therapeutic options evaluated, in the setting of daily clinical practice. Besides, we must not forget that bevacizumab presents a series of important limitations that are usually frequent in this type of patients, because it is contraindicated in haemoptysis cases, or when the tumour is adjacent to a major thoracic vessel. Meanwhile, according to the outcomes of this model, the conclusion that can be reached is that the cis/pem treatment is cost-effective, and that the launch of generic pemetrexed could turn it into the dominant 1st line treatment option for NSCLC patients.


Ferlay J, Soerjomataram I, Ervik M, Dikshit R, Eser S, et al. GLOBOCAN 2012 v1.0, Cancer Incidence and Mortality Worldwide: IARC. [ Links ]

D’Addario G, Fruh M, Reck M, Baumann P, Klepetko W, et al. Metastatic non-small-cell lung cancer: ESMO clinical practice guidelines for diagnosis, treatment and follow-up. Ann Oncol 2010;21(Suppl. 5):v116-9. [ Links ]

National Cancer Institute. SEER stat fact sheets: lung and bronchus. National Cancer Institute; 2012. Available at: [accessed 01.10.12] [ Links ]

Howlader N, Noone AM, Krapcho M, Garshell J, Neymar N, et al. Cancer statistics review, 1975-2011. Bethesda, MD; 2014. [ Links ]

Rapp E, Pater JL, Willan A, Cormier Y, Murray N, Evans WK, Hodson DI, et al. Chemotherapy can prolong survival in patients with advanced non-small-cell lung cancer-report of a Canadian multicenter randomized trial. J Clin Oncol. 1988 Apr;6(4):633-41. [ Links ]

2016 NCCN Guidelines for Non-Small Cell Lung Cancer. [ Links ]

Paz-Ares L, de Marinis F, Dediu M, Thomas M, Pujol JL, et al. Maintenance therapy with pemetrexed plus best supportive care versus placebo plus best supportive care after induction therapy with pemetrexed plus cisplatin for advanced non-squamous non-small-cell lung cancer (PARAMOUNT): a double-blind, phase 3, randomised controlled trial. Lancet Oncol. 2012 Mar;13(3):247-55. [ Links ]

Reck M, von Pawel J, Zatloukal P, Ramlau R, Gorbounova V, et al. Phase III trial of cisplatin plus gemcitabine with either placebo or bevacizumab as first-line therapy for nonsquamous non-small-cell lung cancer: AVAil. J Clin Oncol. 2009 Mar 10;27(8):1227-34. [ Links ]

Sandler A, Gray R, Perry MC, Brahmer J, Schiller JH, et al. Paclitaxel carboplatin alone or with bevacizumab for non-small-cell lung cancer. N Engl J Med. 2006 Dec 14;355(24):2542-50. Erratum in: N Engl J Med. 2007 Jan 18;356(3):318. [ Links ]

Real Decreto-ley 4/2010, de 26 de marzo, de racionalización del gasto farmacéutico con cargo al Sistema Nacional de Salud. BOE de 27 de marzo de 2010;75. Disponible en: 2010-5030.pdfLinks ]

Ministerio de Sanidad, Servicios Sociales e Igualdad. Notas Informativas. Relación informativa de medicamentos afectados por las deducciones establecidas en el Real Decreto Ley 8/2010 de 20 de Mayo por el que se adoptan medidas extraordinarias para la reducción del déficit público. Listado de medicamentos afectados (Octubre 2012). Disponible en: URL: ]

Ministerio de Sanidad, Servicios Sociales e Igualdad. Atlas de la Sanidad en España. Talla y peso medio según grupo de edad y sexo. Disponible en: ]

Dooley MJ, Poole SG, Rischin D, Webster LK. Dosificación de carboplatino: sesgo por sexos y estimaciones inexactas del porcentaje de filtración glomerular. European Journal of Cancer (Ed. Española) 2002;2:125-32. [ Links ]

Parsonage M, Neuburger H. Discounting and health benefits. Health Econ. 1992;1(1):71-6. [ Links ]

Hillman AL, Kim MS. Economic decision making in healthcare. A standard approach to discounting health outcomes. Pharmacoeconomics 1995; 7: 198-205. [ Links ]

Foster TS, Miller JD, Boye ME, Blieden MB, Gidwani R, Russell MW. The economic burden of metastatic breast cancer: a systematic review of literature from developed countries. Cancer Treat Rev. 2011;37(6):405-15. [ Links ]

Oyagüez I, Frías C, Seguí MÁ, Gómez-Barrera M, Casado MÁ, Queralt Gorgas M. Efficiency of oncologic treatments for solid tumours in Spain. Farm Hosp. 2013 May-Jun;37(3):240-59 [ Links ]

Received: October 29, 2015; Accepted: September 29, 2016

* Autor para correspondencia. Correo electrónico: (Fernando Gutiérrez Nicolás).

Conflict of interests

The authors declare that there is no conflict of interests.

Creative Commons License This is an open-access article distributed under the terms of the Creative Commons Attribution License