You are here
Prostate-specific antigen flare induced by cabazitaxel-based chemotherapy in patients with metastatic castration-resistant prostate cancer
European Journal of Cancer, 9, 50, pages 1602 - 1609
A prostate-specific antigen (PSA) flare occurs in about 15% of metastatic castration-resistant prostate cancer (mCRPC) patients receiving docetaxel. This flare has no standard definition. Its impact on treatment efficacy is unclear. We sought to evaluate the incidence and characteristics of PSA flare on cabazitaxel, and its impact on survival.
Multicentre retrospective review of consecutive patients treated with cabazitaxel second-line chemotherapy for mCRPC. Collection of baseline characteristics, disease history and PSA levels before and during cabazitaxel therapy. Overall survival (OS) and radiological/clinical progression-free survival (PFS) for patient groups corresponding to different definitions of PSA flare estimated by the Kaplan–Meier method and compared using the log-rank test.
Overall, 125 patients were included. Median PFS and OS were 6.5 and 13.3 months, respectively. Depending upon the definition used, flare incidence ranged from 8.3% to 30.6%. The flare lasted <2.6 months. A PSA flare followed by a ⩾50% decrease was associated with a median PFS and OS of 11.2 and 25.2 months, respectively. Median PFS and OS for a ⩾30% rather than ⩾50% decrease were 10.4 and 16.5 months. These outcomes were not significantly different from those in patients with immediate PSA decreases of ⩾50% or ⩾30% from baseline, but were significantly better than in patients experiencing no PSA decrease (p = 0.006 and 0.015, respectively, for OS).
The PSA response to cabazitaxel, with or without initial flare, was associated with a strong survival benefit. The taxane-induced flare during the first 12 weeks of therapy can be ignored when evaluating PSA response.
Keywords: Castration resistant prostate cancer, Cabazitaxel, Prostate-specific antigen, Chemotherapy, Cytotoxic agents, Survival rate.
Over the last 10 years or so, major advances have been made in the treatment of metastatic castration-resistant prostate cancer (mCRPC)  . Cabazitaxel is a newly available taxane which has shown survival benefit in mCRPC patients progressing during or after docetaxel therapy, , , , and . As for any other drug, clinicians need to learn to use the drug properly to optimise patient outcomes. A specific issue that needs to be addressed with taxanes is the early PSA flare. This is an isolated PSA rise that occurs after initiation of taxane therapy. It is uncertain whether it should lead to treatment discontinuation or not.
An early PSA flare (aka ‘PSA surge’) followed by a decrease in PSA was first reported in 2004  and has been documented in up to 20% of mCRPC patients treated with first-line docetaxel chemotherapy but its impact on treatment outcome and patient management is unclear. Four retrospective studies have concluded that such a flare is not associated with adverse outcomes and can therefore be disregarded (, , , and , see  for review). However, these studies involved small patient numbers and did not use a standard definition for the PSA flare. Any initial rise, whether relative or absolute, in PSA is considered to signal a flare. On the other hand, there is no consensus on the extent of the subsequent PSA decline for the phenomenon to be a flare. It may be a ⩾50% decline of the peak or baseline PSA value, a stable PSA of no given threshold, or even an undefined PSA response, , , and . In addition, available studies do not indicate clearly whether the observed lack of impact of the flare on treatment outcomes applies to just docetaxel, to just first-line taxane chemotherapy, or can be generalised to various chemotherapy regimens.
The aim of our study was to establish the incidence and characteristics of the PSA flare in mCRPC patients treated with second-line cabazitaxel after progression during or after docetaxel therapy. We used different definitions of PSA flare to assess the impact of flare on overall survival (OS) and on clinical or radiological progression-free survival (PFS). The underlying purpose was to determine whether the flare should or should not prompt treatment discontinuation in clinical practice.
2. Patients and methods
Data on consecutive mCRPC patients treated with cabazitaxel for progression during or after docetaxel were collected retrospectively in nine centres (eight in France and one in Turkey) using an electronic case report form. Patients had received cabazitaxel between January 2007 and May 2012 either in the TROPIC phase III trial ( NCT00417079 ) or in the cabazitaxel compassionate use programme implemented before the availability of the drug on the market. Cabazitaxel was administered as recommended, i.e. 25 mg/m2(or exceptionally 20 mg/m2at the physician’s discretion) every 3 weeks in association with 10 mg daily prednisone or equivalent, with proper anti-allergic/anti-emetic premedication and granular colony-stimulating factor support. Treatment was pursued until disease progression, unacceptable toxicity, or a maximum of 10 cycles in the clinical trial.
For inclusion in the present study, patients had to have received at least two cycles of cabazitaxel treatment and, if alive, had to have undergone at least 6 months follow-up. Patients’ baseline characteristics, detailed disease history (including prior therapies), and all PSA measurements performed before and during cabazitaxel therapy were collected in 2012 for pooled analysis in December 2012. PSA was measured at each treatment cycle in TROPIC trial participants and in nearly all patients of the compassionate use programme. Imaging (bone scan with or without computerized tomography (CT)-scan for patients with measurable disease) was performed every two or three cycles at the physician’s discretion.
We determined the incidence, amplitude and time to peak for six different definitions of PSA flare. A flare was defined as any rise in PSA followed by a decrease which could be: (1) any decrease, (2) any decrease below baseline, (3) a ⩾50% decrease from the peak PSA value, (4) a ⩾50% decrease from baseline, (5) a ⩾30% decrease from the peak value, and (6) a ⩾30% decrease from baseline. In each instance, we determined OS from the first day of cabazitaxel therapy to death and radiological and/or clinical PFS according to the recommendations of Prostate Cancer Working Group 2 (PCWG2)  .
On statistical testing, a two-sidedpvalue of 0.05 was considered significant. OS and PFS were estimated by the Kaplan Meier method and were compared between groups using the log-rank test. Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated with a Cox proportional hazards model. A landmark analysis was performed at 3 months from the first cabazitaxel cycle to remove guarantee-time bias. We used SAS® software (SAS Institute, NC, Cary, United States of America [USA]).
This retrospective non-interventional observational study was approved by the Ethics Committee. Only confidentiality approval was required and was obtained from theCommission nationale de l’informatique et des libertés(CNIL). Informed consent was not required as data were anonymised.
3.1. Population characteristics and outcomes of cabazitaxel therapy
At the time of data collection, 170 mCRPC patients had been treated with cabazitaxel in the participating centres. Twelve patients were excluded because they had received only one cycle of cabazitaxel and 33 patients because of insufficient follow-up. This left 125 patients for inclusion in the present study (46 from the TROPIC trial; 79 from the compassionate use programme). Their characteristics, including data on prior treatments, are given in Table 1 . Median age at the first cycle of cabazitaxel treatment was 67 years. Median follow-up of the patients who were still alive at the time of data collection was 14.3 months (range, 6–59). All deceased patients had undergone complete follow-up until death. The median number of cabazitaxel treatment cycles was 6 (range, 2–14). No patient discontinued cabazitaxel for an isolated PSA rise before the fourth cycle of therapy. Details on treatment and outcomes are given in Table 2 . A PSA decrease of ⩾50% and ⩾30% was achieved in 41.3% and 48.8% of patients, respectively. Median OS and clinical or radiological PFS were 13.3 and 6.5 months, respectively.
Unless otherwise specified.
mCRPC, metastatic castration-resistant prostate cancer; ECOG, Eastern Cooperative Oncology Group; PSA, prostate specific antigen.
Unless otherwise specified.
G-CSF, granulocyte colony-stimulating factor; PSA, prostate specific antigen; CI, confidence interval; OS, overall survival; PFS, progression-free survival.
3.2. Outcomes according to definition of PSA flare
The incidence of the PSA flare, its relative amplitude and the time to PSA peak, as well as subsequent survival (PFS and OS), varied according to the definition used for flare (definitions 1–6) ( Table 3 ). Flare incidence ranged from 8.3% to 30.6%, the lowest incidence being recorded for the most stringent definition (definition 4: flare followed by a ⩾50% decrease from the baseline value). The median PSA peak (percentage rise from baseline) was lowest (19.8%) and median time to peak was shortest (0.6 months, range 0.2–1.4 months) for this stringent definition. However, sample size was smallest. Median time to reach the nadir for definition 4 was 4.2 months (range, 2.0–7.4 months). A flare could last for up to 2.6 months with the least stringent definitions (definitions 1 and 2).
|PSA response||Reference for definition||Patients N (%)||Median PSA rise from baseline (%)||Median time to peak PSA [range] (mos)||Median PFS (mos)||Median OS (mos)|
|Flare + decline|
|1||Any decline from peak||||37 (30.6)||38.7||1.2 [0.2–2.6]||6.9||13.3|
|2||Any decline below baseline||–||25 (20.7)||26.6||0.8 [0.2–2.6]||8.0||16.5|
|3||⩾50% decline from peak and below baseline|| and ||16 (13.2)||39.8||0.8 [0.2–1.9]||10.1||16.5|
|4||⩾50% decline from baseline||||10 (8.3)||19.8||0.6 [0.2–1.4]||11.2||25.2|
|5||⩾30% decline from peak, and below baseline||–||22 (18.2)||38.7||0.8 [0.2–2.1]||8.0||16.5|
|6||⩾30% decline from baseline||–||14 (11.6)||32.6||0.7 [0.2–1.4]||10.4||16.5|
|Immediate decline from baseline|
|⩾50% decline||40 (33.1)||NA||NA||8.8||20.1|
|⩾30% decline||45 (37.2)||NA||NA||9.0||20.1|
PSA, prostate specific antigen; CI, confidence interval; PFS, progression-free survival; OS, overall survival; NA, not applicable.
Four patients were not properly evaluable for PSA response.
Patients experiencing a flare with a ⩾50% decrease from baseline (definition 4) had an estimated median OS and PFS that did not differ significantly from those of immediate responders (OS: 25.2 versus 20.1 months,p = 0.46; PFS: 11.2 versus 8.8 months,p = 0.96). However, their median OS and PFS were significantly better than those of non-responders (OS: 25.2 versus 10.5 months,p = 0.006; PFS: 11.2 versus 3.0 months,p = 0.003) ( Table 4 , Figs. 1 A and 2 A). Survival results for patients experiencing a flare with a ⩾30% decline from baseline (definition 6) were similar ( Table 4 , Figs. 1 B and 2 B). These results on survival were confirmed in a landmark analysis at 3 months from the first cabazitaxel cycle (see Supplementary Table S1 ).
p values with respect to flare.
PSA, prostate specific antigen; HR, hazard ratio; CI, confidence interval.
Our study analysed the early PSA flare occurring on second-line chemotherapy of mCRPC patients with the recently available taxane, cabazitaxel. Two-thirds of the 125 patients had been treated outside a clinical trial in a real-life setting. The incidence of a PSA flare ranged from 8.3% to 30.6% depending upon the definition used for flare and was roughly similar to that recorded for first-line docetaxel chemotherapy (11–18%), , , and . The time to peak of the cabazitaxel-induced PSA flare was also similar to that reported for a docetaxel-induced flare (0.2–2.6 months versus 0.2–2 months  ).
When defined as a rise followed by a decrease of at least 50% or 30% from baseline (definitions 4 and 6), a cabazitaxel-induced PSA flare was associated with a median PFS and OS that were not significantly different from the OS and PFS recorded in patients experiencing an immediate PSA response. Absence of a significant difference had already been observed with first-line docetaxel. On the other hand, survival (whether OS or PFS) of patients with no PSA decrease during cabazitaxel therapy was significantly worse.
Although our study has explored the association between the definition for a PSA flare and survival, it has not definitively resolved the debate on the choice of definition. We consider that, for an accurate comparison with immediate responders, it is preferable to calculate PSA response from the baseline and not peak PSA value. However, whether a decrease of ⩾50% or ⩾30% from baseline is preferable remains a moot point. A PSA decrease of ⩾30% within 3 months of treatment provided the highest degree of surrogacy for survival in the TAX327 trial  whereas the stricter criterion of a ⩾50% decrease is easier and faster to calculate mentally in daily clinical practice.
Our overall PSA response rates and survival times (median OS and clinical or radiological PFS of 13.3 and 6.5 months, respectively) support published outcomes for second-line cabazitaxel therapy in mCRPC patients  . They also support a broad relationship between survival and the magnitude of the PSA decline, as suggested by an ancillary study of a large-scale phase III trial on docetaxel and . The PSA decline may be due to inhibition of androgen receptor (AR) nuclear translocation. Growing evidence shows that taxanes impair cell trafficking and inhibit AR nuclear translocation by stabilizing spindle microtubules, , and . They inhibit both ligand-dependent and ligand-independent transcriptional activities; inhibition might also be mediated by the nuclear accumulation of forkhead box protein O1 (FOXO1), a potent AR-suppressive nuclear factor  . In the case of the well-known luteinizing–hormone–releasing hormone (LH–RH) analogue-induced PSA and clinical flare, initial pituitary hyper-stimulation by the analogue leads to transient hypertestosteronaemia and increased transcription of AR target genes, such as the PSA gene (KLK3)  . Possible mechanisms for the taxane-induced flare may be PSA release from lytic tumour cells, heterogeneous tumour sensitivity to taxanes, or transactivation of a ‘promiscuous’ AR by corticosteroid premedication, , , and .
Our findings on PSA flare and survival have important implications for routine clinical practice because they confirm that cabazitaxel should not be withdrawn in mCRPC patients experiencing an early isolated PSA rise. This is especially relevant in a second-line chemotherapy setting where physicians need to take quick decisions because the disease is progressing rapidly, patient anxiety is growing and treatment costs need to be considered. However, even if an early isolated rise should not be a reason for discontinuing taxane therapy, we recommend not foregoing PSA monitoring during the first cycles of therapy because any early response that may occur might be a useful criterion in evaluating the risk–benefit ratio of therapy. Monitoring PSA is a useful inexpensive means of predicting response even if PSA response probably explains only part of the efficacy of taxanes. New monitoring tools are currently being developed (e.g. circulating tumour cell or circulating DNA counts and analysis), , , and and new therapies such as bone metastasis-directed cabozantinib seem to be effective in mCRPC without inducing any PSA response  .
The main limitation of this study conducted in consecutive patients in a multicenter setting was its retrospective design. Other limitations are some lack of standardization in patient follow-up between the clinical trial and real-life settings. The combination of the two patient populations, however, allowed us to reach a critical population size which, to our knowledge, is the largest so far in a study addressing taxane-induced PSA flare in mCRPC patients.
In conclusion, the initial PSA flare during second-line cabazitaxel therapy was not associated with a worse outcome if followed by a PSA decrease of ⩾30% or ⩾50% from baseline. This finding provides additional support to the PCWG2 recommendation that an early PSA rise (prior to 12 weeks) on administration of a cytotoxic agent should be ignored in evaluating PSA response and should not be interpreted by clinicians as immediate disease progression necessitating treatment discontinuation  . Moreover, our study highlights the need for consensual definitions and thresholds for PSA-related issues to ensure consistency among studies. A standard definition for PSA flare, from the PCWG for instance, would be welcome.
Conflict of interest statement
- AA: Honoraria: Entity: Sanofi-Aventis, Relationship: Myself.
- AF: Honoraria: Entity: Sanofi-Aventis, Astellas, Janssen, Ferring, Relationship: Myself.
- MO: Consultant or Advisory Role, Entity: Sanofi-Aventis, Janssen, Astellas, Relationship: Myself, Compensation: Compensated.
- FM: Consultant or Advisory Role, Entity: Sanofi-Aventis, GlaxoSmithKline, Bristol Myers Squibb, Celgene, Boston Scientific, Relationship: Myself, Compensation: Compensated.
- GG: Consultant or Advisory Role, Entity: Sanofi-Aventis, Relationship: Myself, Compensation: Uncompensated.
- PB: Honoraria, Consultant or Advisory Role, Entity: Sanofi-Aventis, Janssen, Astellas, Ipsen, Relationship: Myself.
- CM: Consultant or Advisory Role, Entity: Sanofi-Aventis, Relationship: Myself, Compensation: Compensated, Entity: Janssen, Relationship: Myself, Compensation: Compensated, Honoraria, Entity: Sanofi-Aventis, Relationship: Myself, Entity: Janssen, Relationship: Myself.
- KF: Consultant or Advisory Role; Entity: Astellas, Relationship: Myself, compensation: Compensated, Entity: Janssen, Relationship: Myself, Compensation: Compensated, Entity: Sanofi-Aventis, Relationship: Myself, Compensation: Compensated.
- TLMR: Consultant or Advisory Role, Entity: Sanofi-Aventis, Relationship: Myself, Compensation: Compensated, Honoraria, Entity: Sanofi-Aventis, Relationship: Myself, Research funding, Entity: Astellas, Millenium, Relationship: Myself.
- SO: Consultant or Advisory Role; Entity: Sanofi, Novartis, Roche, Bayer, Keocyt, Amgen, Relationship: Myself, compensation: Compensated, Honoraria, Entity: Sanofi, Novartis, Roche, Bayer, Keocyt, Amgen, Pfizer, Relationship: Myself.
- All remaining authors have declared no conflicts of interest.
Guillaume Moriceau, resident oncologist in Saint-Etienne, and all staff who helped collect data in the participating centers.
-  S. Bracarda, C. Logothetis, C.N. Sternberg, S. Oudard. Current and emerging treatment modalities for metastatic castration-resistant prostate cancer. BJU Int. 2011;107:13-20 Crossref
-  A.C. Mita, L.J. Denis, E.K. Rowinsky, J.S. Debono, A.D. Goetz, L. Ochoa, et al. Phase I and pharmacokinetic study of XRP6258 (RPR 116258A), a novel taxane, administered as a 1-hour infusion every 3 weeks in patients with advanced solid tumors. Clin Cancer Res Off J Am Assoc Cancer Res. 2009;15(2):723-730 Crossref
-  P. Vrignaud, D. Sémiond, P. Lejeune, H. Bouchard, L. Calvet, C. Combeau, et al. Preclinical antitumor activity of cabazitaxel, a semi-synthetic taxane active in taxane-resistant tumors. Clin Cancer Res. 2013;19(11):2973-2983 Crossref
-  J.S. De Bono, S. Oudard, M. Ozguroglu, S. Hansen, J.-P. Machiels, I. Kocak, et al. Prednisone plus cabazitaxel or mitoxantrone for metastatic castration-resistant prostate cancer progressing after docetaxel treatment: a randomised open-label trial. Lancet. 2010;376(9747):1147-1154 Crossref
-  T.A. Yap, C.J. Pezaro, J.S. de Bono. Cabazitaxel in metastatic castration-resistant prostate cancer. Expert Rev Anticancer Ther. 2012;12(9):1129-1136 Crossref
-  Y. Loriot, K. Fizazi. Taxanes: still a major weapon in the armamentarium against prostate cancer. Eur Urol. 2013;63(6):983-985 Crossref
-  K. Fizazi, R. Thuret, C. Theodore, D. Prapotnich, M. Di Palma, B. Escudier, et al. Early PSA rise subsequently followed by PSA decline is a common event in prostate cancer patients responders to chemotherapy. Ann Oncol. 2004;15(Suppl. 3) iii114 427P
-  P.J. Olbert, A. Hegele, P. Kraeuter, A. Heidenreich, R. Hofmann, A.J. Schrader. Clinical significance of a prostate-specific antigen flare phenomenon in patients with hormone-refractory prostate cancer receiving docetaxel. Anticancer Drugs. 2006;17(8):993-996 Crossref
-  T. Nelius, T. Klatte, W. de Riese, S. Filleur. Impact of PSA flare-up in patients with hormone-refractory prostate cancer undergoing chemotherapy. Int Urol Nephrol. 2008;40(1):97-104 Crossref
-  R. Thuret, C. Massard, M. Gross-Goupil, B. Escudier, M. Di Palma, A. Bossi, et al. The postchemotherapy PSA surge syndrome. Ann Oncol Off J Eur Soc Med Oncol ESMO. 2008;19(7):1308-1311 Crossref
-  A. Sella, C.N. Sternberg, I. Skoneczna, S. Kovel. Prostate-specific antigen flare phenomenon with docetaxel-based chemotherapy in patients with androgen-independent prostate cancer. BJU Int. 2008;102(11):1607-1609 Crossref
-  T. Nelius, S. Filleur. PSA surge/flare-up in patients with castration-refractory prostate cancer during the initial phase of chemotherapy. Prostate. 2009;69(16):1802-1807 Crossref
-  H.I. Scher, S. Halabi, I. Tannock, M. Morris, C.N. Sternberg, M.A. Carducci, et al. Design and end points of clinical trials for patients with progressive prostate cancer and castrate levels of testosterone: recommendations of the Prostate Cancer Clinical Trials Working Group. J Clin Oncol Off J Am Soc Clin Oncol. 2008;26(7):1148-1159 Crossref
-  A.J. Armstrong, E. Garrett-Mayer, Y.-C. Ou Yang, M.A. Carducci, I. Tannock, R. de Wit, et al. Prostate-specific antigen and pain surrogacy analysis in metastatic hormone-refractory prostate cancer. J Clin Oncol Off J Am Soc Clin Oncol. 2007;25(25):3965-3970 Crossref
-  I.F. Tannock, R. De Wit, W.R. Berry, J. Horti, A. Pluzanska, K.N. Chi, et al. Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med. 2004;351(15):1502-1512 Crossref
-  M.-L. Zhu, C.M. Horbinski, M. Garzotto, D.Z. Qian, T.M. Beer, N. Kyprianou. Tubulin-targeting chemotherapy impairs androgen receptor activity in prostate cancer. Cancer Res. 2010;70(20):7992-8002 Crossref
-  M.S. Darshan, M.S. Loftus, M. Thadani-Mulero, B.P. Levy, D. Escuin, X.K. Zhou, et al. Taxane-induced blockade to nuclear accumulation of the androgen receptor predicts clinical responses in metastatic prostate cancer. Cancer Res. 2011;71(18):6019-6029 Crossref
-  M. Thadani-Mulero, D.M. Nanus, P. Giannakakou. Androgen receptor on the move: boarding the microtubule expressway to the nucleus. Cancer Res. 2012;72(18):4611-4615 Crossref
-  L. Gan, S. Chen, Y. Wang, A. Watahiki, L. Bohrer, Z. Sun, et al. Inhibition of the androgen receptor as a novel mechanism of taxol chemotherapy in prostate cancer. Cancer Res. 2009;69(21):8386-8394 Crossref
-  C. Mahler. Is disease flare a problem?. Cancer. 1993;72(12 Suppl.):3799-3802 Crossref
-  A.Y. Liu, L.D. True, L. LaTray, W.J. Ellis, R.L. Vessella, P.H. Lange, et al. Analysis and sorting of prostate cancer cell types by flow cytometry. Prostate. 1999;40(3):192-199 Crossref
-  H.G. Van der Poel, G.O. Oosterhof, H.E. Schaafsma, F.M. Debruyne, J.A. Schalken. Intratumoral nuclear morphologic heterogeneity in prostate cancer. Urology. 1997;49(4):652-657 Crossref
-  B.J. Feldman, D. Feldman. The development of androgen-independent prostate cancer. Nat Rev Cancer. 2001;1(1):34-45 Crossref
-  S. Halabi, A.J. Armstrong, O. Sartor, J.S. de Bono, E.B. Kaplan, E.J. Small. Prostate-specific antigen decline (PSA) as a surrogate for overall survival (OS) in patients (pts) with metastatic castrate-resistant prostate cancer (mCRPC) who failed first-line chemotherapy. J Clin Oncol. 2012;30(Suppl.) [abstr. 4515]
-  J.S. De Bono, H.I. Scher, R.B. Montgomery, C. Parker, M.C. Miller, H. Tissing, et al. Circulating tumor cells predict survival benefit from treatment in metastatic castration-resistant prostate cancer. Clin Cancer Res Off J Am Assoc Cancer Res. 2008;14(19):6302-6309 Crossref
-  H.I. Scher, X. Jia, J.S. de Bono, M. Fleisher, K.J. Pienta, D. Raghavan, et al. Circulating tumour cells as prognostic markers in progressive, castration-resistant prostate cancer: a reanalysis of IMMC38 trial data. Lancet Oncol. 2009;10(3):233-239 Crossref
-  S. Kwee, M.-A. Song, I. Cheng, L. Loo, M. Tiirikainen. Measurement of circulating cell-free DNA in relation to 18F-fluorocholine PET/CT imaging in chemotherapy-treated advanced prostate cancer. Clin Transl Sci. 2012;5(1):65-70 Crossref
-  D.C. Smith, M.R. Smith, C. Sweeney, A.A. Elfiky, C. Logothetis, P.G. Corn, et al. Cabozantinib in patients with advanced prostate cancer: results of a phase II randomized discontinuation trial. J Clin Oncol Off J Am Soc Clin Oncol. 2013;31(4):412-419 Crossref
a Department of Medical Oncology, Georges Pompidou European Hospital, Paris, France
b Department of Medicine, Centre Léon Bérard, Lyon, France
c Department of Internal Medicine, Istanbul University Hospitals, Istanbul, Turkey
d Stat Process, Port-Mort, France
e Department of Medicine, Institut de Cancérologie de la Loire, Saint-Etienne, France
f Department of Medical Oncology, Val de Grace Hospital, Paris, France
g Department of Medicine, Institut Paoli Calmettes, Marseille, France
h Department of Medicine, Institut Curie, Paris, France
i Department of Cancer Medicine, Institut Gustave Roussy, University of Paris Sud, Villejuif, France
j Department of Medical Oncology, Pitié-Salpêtrière Hospital, Paris, France
© 2014 Published by Elsevier B.V.