This report provides data on the use of a combination of cisplatin and vinorelbine in chemotherapy-naive patients having inoperable MPM. The response rate of 29.6% is noteworthy, being comparable to the most used regimens in MPM (). It is, however, not possible to draw firm conclusions concerning major differences in activity between these regimens for several reasons. First, the majority are either phase II trials or retrospectively analyses except for the randomised trials by Vogelzang et al, 2003
, and van Meerbeeck et al, 2005
, which may not be entirely comparable as there is usually a tendency towards higher response rates observed in the very selective non-randomised trials. Another source of variability is the different mixture of various prognostic variables between the trials. Both analyses of prognostic variables by CALGB in 337 patients (Herndon et al, 1998
), by EORTC in 204 patients (Curran et al, 1998
), and in the randomised trial by van Meerbeeck et al
in 250 patients pointed towards a better outcome for MPM patients having the epithelial histological subtype. A worse prognosis was reported for patients in poor performance status or advanced age. Independent predictors of survival in CALGB study were performance status, age, chest pain, weight loss, leukocyte count, and haemoglobin level, whereas they were histology, performance status, gender, and leukocyte count in the EORTC study. The frequency of epithelial subtype in the current study was 74%, which is somewhat higher than the around 60% usually considered average in the entire MPM population (Vogelzang, 2002
), thus possibly contributing to a tendency towards a higher response rate. The frequency of epithelial subtype was, however, similar to the frequencies in newer studies, such as those by Vogelzang et al, 2003
(68% epithelial), van Meerbeeck et al, 2005
(75%), Berghmans et al, 2005
(74%), and van Haarst et al, 2002
(81%), pointing toward a rather uniform rate of epithelial subtype of 70–80% in more recent trials for chemotherapy in MPM and with the current study being in accordance with these trials. Interestingly, the activity is also comparable to that of single-agent Vinorelbine, which revealed a 24% partial response rate (Steele et al, 2000
Measurement of tumour response to antineoplastic chemotherapy is notoriously difficult both when using the WHO and the RECIST criteria due to the parietal growth patterns of these tumours (Monetti et al, 2004
; van Klaveren et al, 2004
). Recently, modified RECIST criteria for use in MPM have been suggested by Byrne and Nowak, 2004
, measuring tumour thickness perpendicular to the chest wall or mediastinum in two positions at three separate levels on thoracic CT scans. The use of these different systems in various reports may contribute to variations in the response rates observed, in addition to the variations caused by different distributions of known and unknown prognostic variables and possible differences in activity between the treatment regimens under evaluation. There is a growing evidence that therapy-induced changes in tumour FDG uptake, as measured by FDG-PET imaging, might predict response and patient outcome (Ceresoli et al, 2007
). Thus, FDG-PET imaging could be potentially useful in the early assessment of treatment efficacy.
Haematological toxicity was relatively pronounced in the current study of cisplatin and vinorelbine with 48% of patients having leukopenia CTC grade 3 or 4 (). This is, however, in accordance with the 54% grade 3 or 4 leukopenia confined with cisplatin and epirubicin in MPM (Berghmans et al, 2005
) and lower than the 81% of NSCLC patients who had grade 3 or 4 granulocytopenia in the randomised SWOG trial using same doses of cisplatin and vinorelbine as the current study (Wozniak et al, 1998
). It is, however, considerably higher than was observed in the randomised trials in MPM patients with cisplatin and pemetrexed (18%, Vogelzang et al, 2003
) or cisplatin and raltitrexed (7%, van Meerbeeck et al, 2005
). Also, the rate of febrile leukopenia was somewhat higher, being 9% compared to 2 and 1%, respectively, in the two randomised trials cited above. There were no septic deaths and no toxic deaths overall, but the febrile lekopenia rate of 9% in this study is not only a potential risk but also an inconvenience for those patients who have to be admitted for intravenous antibiotics, as well as a cost for the health-care system. The use of granulocyte colony-stimulating factor may diminish the problem but was not a part of this protocol. Other toxicities were generally not pronounced ().
The survival of patients with MPM who received cisplatin and vinorelbine in this study was impressive with 31% being alive after 2 years and a median survival of 16.8 months, even though patients with adverse prognostic variables such as performance status 2 and age above 70 years were included (). The use of second-line chemotherapy in 22 patients may possibly have had an impact on survival. However, these figures are small and there are potential selection biases in the choice of second-line treatment or not. Hence, a possible impact cannot be explored or concluded from these data. The survival results compares favourably with those reported on for other active regimens, ranking the combination of cisplatin and vinorelbine among the most active cytotoxic treatments for MPM reported to date (). On the other hand, this regimen is convenient for neither the patients nor for the health-care system because of the weekly administration of intravenous vinorelbine and because of the haematological toxicity encountered. Other regimens in may be more convenient and less troublesome in the palliative treatment setting. A number of regimens seem to possess similar activity without any regimen being clearly superior. No combination chemotherapy regimens have been compared to each other in randomised trials, but the regimens of both cisplatin and pemetrexed as well as cisplatin and raltitrexed have proved superior to single agent treatment with cisplatin in randomised trials (Vogelzang et al, 2003
; van Meerbeeck et al, 2005
). It is of great importance to use a treatment regimen with a high antitumour activity in another clinical situation such as induction chemotherapy before surgery, and the current regimen may, despite its inconvenience, be among the options for that situation.
Selected combination chemotherapy regimens in MPM
The high activity of cisplatin and vinorelbine deserves attention for use as induction treatment before surgery in resectable cases. Further development of platinum compounds, together with vinorelbine in the palliative setting, also seems justified both in the light of the documented activity of the two-drug combination used in the current study and also because the single agent activity of vinorelbine with a response rate of 24% (Steele et al, 2000
) ranks this agent among the most active single drugs in MPM. It must, however, be kept in mind that these results are obtained from relatively small and non-randomised trials with wide confidence limits. Improvement of the current regimen is necessary if it is to be used in the palliative situation to make it more feasible and carboplatin may thus be used instead of cisplatin and vinorelbine applied in the oral formulation. Further improvement of the regimen in the palliative setting is under evaluation by the investigative group. The oral formulation of vinorelbine may render the regimen more convenient and feasible, and evaluation of new targeted agents is necessary to improve prognosis for this dismal disease. Taking the documented activity of vinorelbine in the first-line treatment of MPM into consideration, vinorelbine may also be explored as second-line treatment for patients not previously exposed to this drug.