Gastrolearning III lezione
Costo efficacia della terapia con sorafenib nel trattamento dell’HCC - Prof. C. Cammà (Università di Palermo)
www.gastrolearning.it
3. BCLC Staging and Treatment Schedule
HCC
Stage A-C Stage D
Stage 0
Okuda 1-2, PST 0-2, Child-Pugh A- Okuda 3, PST>2, Child-Pugh
PST 0, Child-Pugh A
B C
Very early stage (O) Early stage (A) Intermediate stage (B) Advanced stage (C) Terminal
Single < 2 cm Single or 3 nodules Multinodular, Ps 0 Portal invasion, stage (D)
Carcinoma in situ < 3 cm, PS 0 N1, M1, PS 1-2
Single 3 modules ≤ 3 cm
Portal Associated Portal
Increased
pressure/bilirubin diseases invasion, N1, M1
Normal No Yes
Liver Transplantation
Resection PEI/RF Chemoembolization Sorafenib
(CLT/LDLT)
Curative Treatments (30%) Randomized controlled trials (50%) Symptomatic ttc (20%)
5-yr survival: 50-70% 3 yr survival: 20-40% 1 yr survival: 10-20%
ttc: treatment
Llovet JM et al. J Natl Cancer Inst 2008
4. Multicenter, double blind, placebo-controlled trial
conducted at 121 centers in 21 countries in
Europe, North America, South America, and
Australasia
5. Phase III SHARP Trial:
SHARP trial design
•Multicenter, double blind, placebo-controlled trial conducted at
•121 centers in 21 countries in Europe, North America, South
America, and Australasia
Eligibility criteria
Advanced HCC
Primary endpoints
•Randomization (1:1)
Child–Pugh A status • Nexavar®
ECOG PS 0–2 400 mg b.i.d. OS
No prior systemic N=299 TTSP
(n=602)
therapy
Secondary endpoints
Stratification
TTP
Region
•Placebo DCR
ECOG PS
(0 vs 1–2) Safety*
N=303
MVI/EHS
(present/absent)
•ECOG PS = Eastern Cooperative Oncology Group Performance Status; MVI = macroscopic vascular invasion;
EHS = extrahepatic spread; BID = twice daily; OS = overall survival; TTSP = time to symptomatic progression;
TTP = time to progression; DCR = disease control rate
*Assessed using version 3.0 of the USA National Cancer Institute
Common Terminology Criteria for Adverse Events
•Llovet JM et al. ASCO Annual Meeting 2007; Abstr LBA1/oral presentation available at www.asco.org
6. The SHARP Trial: Overall Survival
1.00
Sorafenib
Median: 10.7 months
Probability of Survival
(95% CI: 9.4-13.3)
0.75
Placebo
Median: 7.9 months
(95% CI: 6.8-9.1)
0.50
0.25
HR (S/P): 0.69 (95% CI: 0.55-0.87)
P < .001
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Months Since Randomization
Llovet JM, et al. N Engl J Med. 2008;359:378-390.
7. THE CASE OF HEPATOCELLULAR CARCINOMA
1) Biologica
Marcata 2) Epidemiologica
eterogeneità
3) Diagnostica
4) Clinica
8. Untreated control groups of 30 RCTs
p for heterogeneity < 0.0001 7.3%
17.5%
Range 0 – 75% Range 0 – 50%
Cabibbo et al. Hepatology 2010
9. Safety
• The quality and quantity of safety reporting in
randomized trials are largely inadequate (1);
• RCTs often fail to detect infrequent but serious
adverse drug reactions (1);
• Overall, 50% of approved drugs have serious
adverse effects not detected prior to approval (2).
1. Ioannidis JPA, Lau J. JAMA 2001
2. Moore TJ & al. JAMA 1998
10. 143 RCTs stopped early for benefit
RCTs can overestimate the magnitude of the treatment effect
depending on the timing (ie, expected number of events) of
the decision to stop.
Lack of adequate safety data may in turn affect the perceived
and actual risk-benefit ratios (overestimating the benefit,
underestimating the risk) of implementing the intervention in
clinical practice.
These considerations suggest that clinicians should view
results of RCTs stopped early for benefit with skepticism.
11. Field-practice study of sorafenib therapy for
hepatocellular carcinoma:
a prospective multicenter study in Italy.
Iavarone M, Cabibbo G, Piscaglia F, Zavaglia C, Grieco A,
Villa E, Cammà C, Colombo M;
on behalf of the SOFIA (SOraFenib Italian Assessment) study group.
Hepatology 2011.
12. PATIENTS AND METHODS
Study Multicenter, prospective, observational study
Centers Milan Policlinico, Palermo, Bologna, Milan
Niguarda, Rome, Modena
Patients Consecutive patients with BCLC-C or BCLC-
B with PD/unsuitable to locoregional therapy
Enrollment July 2008 – July 2010
Treatment Sorafenib 400 mg twice daily
Inclusion - compensated cirrhosis
- ECOG 0-2
13. PATIENTS AND METHODS
Management of HCC AASLD guidelines, 2005
Dose reduction 400 mg once daily
AE (grade 3/clinical judgement)
Discontinuation Radiological or clinical progression
SAE
14. DEMOGRAPHY
BCLC-C BCLC-B Overall
Patients 226 (76%) 70 (24%) 296
Age, yr* 66±10 69±10 67±10
Male 185 (82%) 57 (81%) 242 (82%)
HCV/HBV/alcohol abuse/other 118/45/21/42 34/13/10/13 152/58/31/55
ECOG 0/1/2 89/126/11 70/0/0 159/126/11
Child-Pugh A 196 (87%) 63 (90%) 259 (88%)
Macroscopic vascular invasion 115 (51%) NA 115 (39%)
Extrahepatic spread 104 (46%) NA 104 (35%)
*mean±SD, §upper gastrointestinal endoscopy was performed in 256 patients
15. EFFECTIVENESS
1-yr survival rate 49%
Early radiological response (month 2)*
Complete 2 (1%)
Partial 22 (7%)
Stable 217 (73%)
Progressive disease 55 (19%)
Time to radiological progression, months 9.2 (5.5-12.9)
*according to modified RECIST criteria (Llovet JM et al. J Natl Cancer Inst 2008;100:698-711)
16. Overall survival of patients treated with
sorafenib
(from RCTs to clinical practice)
SHARP trial ° SOFIA study §
1-year survival: 44% 1-year survival: 49%
Sorafenib Arm
Median Survival (n= 299): 10.7 mo Median Survival (n= 296): 10.5 mo
°Llovet JM, et al. N Engl J Med. 2008. §
Iavarone, Cabibbo et al. Hepatology 2011.
17. Overall survival of patients treated with
sorafenib
according to BCLC
SHARP trial °* SOFIA study §
1.00
Probability of Survival
P =ns
0.75
B
0.50
0.25 C
0
0 4 8 12 16
Months
BCLC B (n= 54; 18%) Median: 14.5 mo BCLC B (n= 74; 25%) Median: 20.6 mo
BCLC C (n= 245; 82%) Median: 9.7 mo BCLC C (n= 222; 75%) Median: 8.4 mo
°Llovet JM, et al. N Engl J Med. 2008;
§
Iavarone, Cabibbo et al. Hepatology 2011.
° Bruix J, et al. J Hep 2009; S28.
19. SOFIA
SHARP
Liver dysfunction < 1% Liver function deteriorated 15%
(≥ 2 points of Child-Pugh score)
20. Treatment compliance of patients
treated with sorafenib
(from RCTs to clinical practice)
SHARP trial ° SOFIA study §
Discontinuation due to AEs Discontinuation due to AEs
38% 45%
vs.
Dose reductions due to AEs Dose reductions due to AEs
26% 54%
Dose interruptions due to AEs
Dose interruptions due to AEs
44% 56%
°Llovet JM, et al. N Engl J Med. 2008. §
Iavarone, Cabibbo et al. Hepatology 2011.
21. Ma la riduzione di dose nel corso del trattamento,
inficia l’efficacia della terapia?
22. - 77 (26%) patients received a half-dose of sorafenib for ≥ 70% of
the treatment period, which lasted a median of 6.8 months (95%
CI 4.2-9.4).
- Among the remaining 219 patients (74%),136 maintained a full
dose of sorafenib for a median of 3 months (95% CI 2.2-3.8),
whereas 83 received a half-dose for <70% of the whole treatment
period of 3 months.
23. Survival according to sorafenib dose reduction
(Post-hoc analysis)
--- Full dose No. = 219
--- Half-dose No. = 77
P = 0.0006
Median 21.6 months (95% CI 13.6-29.6) vs
9.6 months (95% CI 6.9-12.3)
§
Iavarone, Cabibbo et al. Hepatology 2011.
24. Predictors of mortality in 296 HCC patients
treated with sorafenib
Multivariate analysis
Predictor HR (95% CI) P-value
ECOG 1.9 (1.5 – 2.5) <.0001
Vascular invasion 1.9 (1.4 – 2.6) 0.0009
Full dose 1.8 (1.4 – 2.4 ) 0.001
Extrahepatic spread 1.4 (1.1 – 1.9) 0.01
Early radiological progression 1.4 (1.1 – 2.1) 0.02
Total bilirubin – mg/dl - -
Platelet x 103/mmc - -
Age - -
Albumin – g/dl - -
§
Iavarone, Cabibbo et al. Hepatology 2011.
25. The open-label, Phase III SUN1170 trial compared the efficacy and safety of sunitinib
with that of sorafenib
Sunitinib
1.00 Median 7.9 months (95% CI: 7.4-9.2)
Sorafenib
Median 10.2 months (95% CI: 8.9-11.4)
OS probability (%)
0.75
HR 1.30 (95% CI: 1.13-1.50)
P=.0010
0.50
0.25
0.0
0 5 10 15 20 25 30 35 40
“The SUN 1170 trial was stopped early because of a higher
incidence of serious adverse events in the sunitinib arm, and
because sunitinib did not demonstrate superiority or non-
inferiority to sorafenib.”
Zhu A, et al. NATURE REVIEWS | CLINICAL ONCOLOGY 2011
26. Ma uno studio pianificato per verificare se un farmaco è
«non peggiore» rispetto ai trattamenti standard, senza
nessun interesse per alcun valore aggiunto, non pone
alcuna domanda clinicamente rilevante.
Inoltre, a meno che non sia già nota una maggiore
tollerabilità del nuovo farmaco, comporta
inevitabilmente un inaccettabile eccesso di eventi
avversi nella popolazione dei pazienti.
27.
28. Survival curves of BCLC B+C patients treated
with full dose or dose-adjusted sorafenib
Kaplan-Meier curves: stair-step line
Estimated curves: smooth line
29. Survival curves of BCLC B patients treated
with full dose or dose-adjusted sorafenib
Kaplan-Meier curves: stair-step line
Estimated curves: smooth line
30. Survival curves of BCLC C patients treated
with full dose or dose-adjusted sorafenib
Kaplan-Meier curves: stair-step line
Estimated curves: smooth line
32. Results of cost-effectiveness analyses
Treatment Strategies according Costs in 2012 QALY ICER/QALY
BCLC and dose euros base-case analysis
(2012 euros)
Full dose for BCLC B and C 16,081 0.16 69,344
Dose-adjusted for BCLC B and C 19,944 0.44 34,534
Full dose for BCLC B 24,224 0.32 57,385
Dose-adjusted for BCLC B 26,914 0.38 54,881
Full dose for BCLC C 14,841 0.16 65,551
Dose-adjusted for BCLC C 16,625 0.44 27,916
39. CONCLUSIONI
EBM ha implementato l’uso delle “evidenze” dalla ricerca clinica
come strumento principale di decision making.
Tuttavia, i problemi maggiori degli RCTs sono:
• i risultati favorevoli ottenuti attraverso disegno, conduzione
ed analisi inadeguati;
• l’eccessiva fiducia nel risultato “medio” nonostante la verosimile
eterogeneità tra pazienti;
• la sottovalutazione di sicurezza e tollerabilità;
• la validità esterna insufficiente;
• l’influenza pervasiva dell’industria.
42. Ciascun individuo è un esperimento
biologico inripetibile, ……………..
nella storia dell’umanità.
Giovanni Paolo II
Editor's Notes
If something ’ s not worth doing then it ’ s not worth doing well!
Lo schema di trattamento BCLC propone differenti opzioni terapeutiche in base allo stadio del tumore. I pazienti in stadio 0 (HCC molto precoce) sono candidati ideali alla resezione. I pazienti in stadio A presentano tumori in stadio precoce asintomatici e sono idonei a trattamenti radicali (resezione, trapianto, trattamenti percutanei, quali alcolizzazione o radiofrequenza). Lo stadio B (HCC intermedio) include pazienti con HCC multinodulare asintomatico che possono trarre beneficio dalla chemioembolizzazione. Nello stadio C (HCC avanzato) sono inclusi pazienti con tumori sintomatici e/o con invasione vascolare/disseminazione extraepatica. Questa categoria di pazienti è idonea al trattamento con sorafenib. I pazienti appartenenti allo stadio D presentano una prognosi molto sfavorevole, per loro si suggerisce un trattamento sintomatico. Llovet JM et al. J Gastroenterol 2005; 40: 225-235; Llovet JM et al. J Natl Cancer Inst 2008;100: 698 – 711
Lo studio di fase III SHARP è uno studio multicentrico, randomizzato, in doppio cieco, controllato con placebo. L ’ endpoint primario era la sopravvivenza globale (OS) e il tempo alla progressione sintomatica (TTSP); gli endpoint secondari erano il tempo alla progressione del tumore (TTP), il controllo della malattia e la sicurezza. Llovet JM et al. ASCO Annual Meeting 2007; Abstr LBA1/oral presentation available at www.asco.org
Come suggerito da Piscaglia, non ho citato direttamente il Child Pugh, bensì una frase generica “compensated cirrhosis. A voce si potrebbe dire che il trapianto di fegato non era un criterio di esclusione. Per il momento ho lasciato prospettico, sarà tema di discussione giovedì o venerdì mattina alla nostra riunione
Ho provato a lasciare la demografia divisa per BCLC, con overall come ultima colonna. Avrei bisogno del valore MELD per sottogruppo (Prof Cammà/Cabibbo)
Questa diapo per dare il dato di sopravvivenza a 12 mesi (overall), ma soprattutto per introdurre i 55 early progressors e il time to radiological progression (più lungo rispetto a SHARP, verosimilmente per l ’uso del mRECIST?)
Qui a voce identificherò i predittori di mortalità alla multivariata (che identificano il BCLC C non lo dirò …)