Controversies in Neoadjuvant Therapy for Esophageal / GEJ Carcinoma

1
Controversies in Neoadjuvant Therapy for
Esophageal / GEJ Carcinoma

• Gordon Buduhan MD MSc FRCSC
• Division of Thoracic Surgery
• Dalhousie University
• Halifax, NS

2
No conflicts of interest
3
Introduction

• Esophageal carcinoma most rapidly increasing
  tumor type in Western world
• Standard of care for resectable esophageal ca
  controversial
• Majority of centers multimodality approach
  including surgery, chemotherapy, radiation
• 2 most common approaches
• Neoadjuvant chemo ? surgery ( post-op chemo)
• Neoadjuvant chemoradiation ? surgery

4
Rationale for Neoadjuvant Therapy

• Treat micrometastatic disease
• Downstage tumor, increase probability of curative
  resection
• Decrease tumor cell dissemination during
  resection
• Patients more likely to complete prescribed
  treatment in preoperative period

5
Ann Surg Oncol 2011
6
Geographic Practice Variations
Neoadjuv ChemoRT
Neoadjuv CT
7
Chemo radiation more better!
8
The argument for neoadjuvant CRT (over CT)

• Better local control
• Improved pathologic CR rate
• No added morbidity / mortality with addition of
  XRT
• Improved survival evidence from RCTs

9
The argument for neoadjuvant CRT (over CT)

• Better local control
• Improved pathologic CR rate
• No added morbidity / mortality with addition of
  XRT
• Improved survival evidence from RCTs

10
The Importance of Complete Resection
Zafirellis K, et al. Dis Esoph, 2002
11
Neoadjuvant chemo can increase R0 resection rate

• 2 cycles preop cis, 5FU vs. surg alone
• N802 pts esoph SCC or ACA
• R0 (microscopically complete) resection rate 60
  in chemo grp (vs. 53 surg alone, p0.0001)
• Significant OS improvement in chemo grp (more
  on this later)

MRC OEO2. Lancet 2002
12
Influence of Type of Surgery After Neoadjuvant
Therapy

• Retrospective comparison - pts that had
  neoadjuvant therapy followed by either
  transhiatal or en bloc esophagectomy at USC from
  1992-2005
• 90 transhiatal and 95 en bloc esophagectomy pts
  neoadjuv CRT
• N 58 pts

Rizzello et al. JTCVS 2008
13
Pathologic Stage After Resection
Transhiatal group (n18)
En bloc group (n40)
pn.s.
Stage 0 Stage I Stage IIA Stage IIB Stage III
Rizzello et al. JTCVS 2008
14
Results
Transhiatal Group (n18) En bloc Group (n40) p value
Complete pathologic response 7 (38.8) 10 (25) 0.35
Pts with residual esophageal cancer 10 (55.5) 29 (72.5) 0.23
Pts with at least 1 involved node 7 (38.8) 19 (47.5) 0.58
Rizzello et al. JTCVS 2008
15
Recurrence Pattern after Esophagectomy
Transhiatal Group (n18) En bloc Group (n40) p value
Any recurrence Systemic Local/regional Neck 13 (68) 9 (50) 3 (16.6) 1 (5.5) 19 (47.5) 18 (45) 0 1 (2.5) 0.095 0.78 0.023 0.51
Rizzello et al. JTCVS 2008
16
Overall 5-yr Survival in all Patients
p0.04
Rizzello et al. JTCVS 2008
17

• Preop cis, 5FU, concurrent XRT 35 Gy vs. surg
  alone
• N256
• R0 resection rate 80 in CRT grp (vs. 59 surg
  alone, p0.003)
• BUT

Burmeister et al. Lancet Oncol 2005
18
Most recurrences following esophagectomy are
systemic, not locoregional
Pattern of Recurrence post Esophageal Cancer Resection () Pattern of Recurrence post Esophageal Cancer Resection () Pattern of Recurrence post Esophageal Cancer Resection ()
Locoregional Hematogenous / distant Mixed
Osugi Oncol Rep 2003 11 58 -
Kato Anticancer Rsrch 2005 22 51 27
Fahn ATS 1994 33 61 -
Abate JACS 2010 30 60 10
19
Bottom line for local control

• Neoadjuvant chemotx alone may improve R0
  resection rate
• Addition of preop XRT to chemo may also improve
  completeness of resection, but not necessarily
  associated with improved DFS / OS
• Most recurrences hematogenous ? systemic
  treatment is the key

20
The argument for neoadjuvant CRT (over CT)

• Better local control
• Improved pathologic CR rate (no residual tumor
  cells)
• No added morbidity / mortality with addition of
  XRT
• Improved survival evidence from RCTs

21
Fact

• PathCR rates consistently higher with neoadjuvant
  CRT (25-40) compared to neoadjuvant CT (2-10)

22
Assumption

• Pathologic response to neoadjuvant therapy is a
  reliable surrogate for clinical outcome i.e.
  survival
• or is it??

23
(No Transcript)
24
Randomized Trial of Preoperative Chemoradiation
Versus Surgery Alone in Patients With
Locoregional Esophageal Carcinoma

• N100
• RCT preop CRT (cis, 5FU, vinblastine 45 Gy) vs.
  surg alone
• 28 pathCR median OS 49.7 mo
• vs.
• Residual disease median OS 12 mo
• P0.01

Urba et al. JCO 2001
25

• N118, neoadjuv CRT ? surgery
• pathCR 32
• No survival advantage in pathCR compared with no
  path CR pts

SSO 57th Annual Cancer Symposium 2004
26
Induction Chemoradiotherapy Followed by
Esophagectomy in Patients With Carcinoma of the
Esophagus

• 54 pts
• Cisplatin, 5FU concurrent XRT ? esophagectomy

NO significant difference 3 yr OS b/w pathCR and
no pathCR grps (log-rank p0.13)
Jones et al. ATS 1997
27
Pathologic Complete Response May Not
Representthe Optimal Surrogate for Survival
AfterPreoperative Therapy for Esophageal Cancer

• Phase II trial - 23 pts
• hyperfractionated XRT twice-daily XRT wk 1 5,
  once-daily XRT wks 2-4 (59 Gy)
• Cis on D1, 5FU 1st 5th wks of XRT
• 19 pts had esophagectomy pCR 16
• study closed _at_ interim analysis having not met
  required minimum pCR rate of 20
• BUT median OS 44.6 mo 65 of pts alive at 2
  yrs

With the encouraging 2-yr survival, it is not
clear that pCR is a reliable surrogate endpoint
to discern treatment efficacy
Blackstock et al. Int J GI Ca 2006
28

• Recurrence and survival after pathologic complete
  response to preoperative therapy followed by
  surgery for gastric or gastrooesophageal
  adenocarcinoma
• N60 pCR post neoadjuvant therapy GEJ / gastric
  adenoca
• Recurrence _at_ 5 yrs lower for pCR vs non-pCR pts
  (27 and 51, P0.01)
• Probability of recurrence for pts with pCR
  similar to non-pCR pts with stage I or II disease
• Pattern of local / regional (LR) vs distant
  recurrence comparable (43 LR vs 57 distant)
  between pCR and non-pCR grps
• ? incidence CNS first recurrences in pCR patients
  (36 vs 4, P0.01)

Fields et al. Br J Cancer 104, 1840-1847 (7 June
2011)
29
Is it really a CR? depends how hard you look for
it!

• N18 pts with pathCR H/E stain post neoadjuv CRT
• IHC CK staining 7/18
• 3 of 7 had microscopic residual disease not
  detected on H/E
• 3/7 CK had systemic relapse

Chadha et al. Int J Oncol 2008
30
Bottom line for pathCR

• Some correlation between pathCR and ?survival,
  but not consistently reported
• pathCR may not be optimal surrogate outcome
• Need to look for it

31
The argument for neoadjuvant CRT (over CT)

• Better local control
• Improved pathologic CR rate
• No added morbidity / mortality with addition of
  XRT
• Improved survival evidence from RCTs

32
Harmful effects of radiation!
33
Increased pulmonary complications with
neoadjuvant chemoradiation esophagectomy

• Avendano et al. ATS 2002
• preop CRT longer duration mechanical ventilation
  c/w preop chemo only (mean 187 vs. 10 hrs)
• Abou-Jawde et al. Chest 2005
• 45 Gy 22 median ? DLCO 17 postop resp
  complications
• Lee et al. Int J Rad Oncol Biol Phys 2003
• Median 45 Gy 18 major pulm complications

34
Increased mortality with neoadjuvant
chemoradiation esophagectomy

• Bossett et al. RCT neoadjuv CRT vs surg alone
  NEJM 1997
• 12 periop mortality in CRT grp (vs. 4 surg
  alone, p0.012) no diff OS b/w grps
• Urschel meta-analysis RCTs preop CRT vs surg Am
  J Surg 2003
• preop CRT trend toward ?operative (p0.07) and
  all-cause mortality (p0.05) c/w
  surg alone
• Urschel meta-analysis RCTs preop CT vs surg Am J
  Surg 2002
• Preop CT no difference operative (p0.76) or
  treatment mortality (p0.22)

35
Increased mortality with neoadjuvant
chemoradiation esophagectomy

• FFCD 9901 (France)
• RCT 195 pts Stage 1, 2 esophageal SCC /
  adenoca surg alone vs. CRT 5-FU 800 mg/m2/day
  D1-4, cisplatin 75 mg/m2 D1/2 x 2 cycles, 45 Gy
  concurrent rad
• - median OS 32 mo CRT arm vs. 44 mo surg arm
  p0.66
• - 30 day mortality 7.3 CRT arm vs. 1.1
  surg p0.05

J Clin Oncol 2815s, 2010
36
BUT

• Many other RCTs have shown no increase in
  post-op mortality post neoadjuvant CRT

Tepper JCO 2008 Van der
Gaast ASCO 2010 Burmeister Lancet Oncol 2005
37
The argument for neoadjuvant CRT (over CT)

• Better local control
• Improved pathologic CR rate
• No added morbidity / mortality with addition of
  XRT
• Improved survival evidence from RCTs

38
Neoadjuvant chemotherapy
39
Intergroup 0113 MRC OEO2
pts 440 802
Chemo regimen Cis 100 mg/m2, 5FU 1000 mg/m2 x 3 cycles preop, x 2 cycles postop Cis 80 mg/m2, 5FU 1000 mg/m2 x 2 cycles preop
Duration chemo 12 wks 6 wks
completed chemo 71 90
Median time (days) randomization - OR 93 63
pts in chemo arm that did not have surgery 20 6.5
40
MRC OEO2 trial long term F/U

• Median F/U 6 yrs
• Recurrences reduced by 18 (P 0.003)
• Mortality reduced by 16 (P 0.03)
• 5 yr OS 23 with neoadjuvant chemotherapy vs. 17
  with surgery alone
• Survival advantage in favor of neoadjuvant
  chemotherapy - consistent in both pts with ACA
  (5-yr OS 24 vs 17) and SCC (23 vs 18)

41
Meta-analysis neoadjuvant chemo vs. surgery in
esophageal ca

• Overall surv benefit (HR 0.90, 95 CI 0.81-1.0
  p0.05)
• 2 yr absolute surv benefit 7 (over surg alone)

Gebski et al. Lancet Oncol 2007
42
MAGIC trial

• Gastric / GEJ adenoca, n503
• 3 cycles ECF pre-op, 3 cycles post-op vs. surg
  only
• 5 yr survival 36 chemo grp (vs. 23 in surg grp)
  p0.009
• Only 26 GEJ / esoph ca BUT no heterogeneity
  treatment effect based on tumor location
• 91 pts completed preop chemo, 50 completed
  postop chemo

43
FNLCC ACCORD07-FFCD 9703 trial

• N224 adenoca (75 GEJ)
• Surgery alone vs. preop chemo cis, 5FU x 2-3
  cycles preop ? surg ? 1-4 cycles postop if pLN
  or response to preop tx
• Median F/U 6 yrs
• R0 resection rate 87 chemo arm (vs. 74 surg
  arm, p0.04)
• 5 yr OS 38 chemo grp (vs. 24 surg, p0.021)

J Clin Oncol, 2007 ASCO Annual Meeting
Proceedings Part I. Vol 25, No. 18S (June 20
Suppl), 2007 4510
44
Neoadjuvant Chemo
MRC OEO2
?
MAGIC
?
FFCD
?
45
Neoadjuvant chemoradiation
46
Meta-analysis neoadjuvant chemorad vs. surgery in
esophageal ca

• Overall survival benefit (HR 0.81 95CI
  0.7-0.93, p0.002)
• 2 yr absolute survival benefit 13 (over surg
  alone)

Gebski et al. Lancet Oncol 2007
47
Flaws in neoadjuvant CRT trials
Walsh
Tepper CALGB 9781

• Single institution, n113
• Surgery alone grp poor survival (6 3 yr OS)
• Staging issues done after completion CRT, used
  CXR, UGI, U/S

• Underpowered!!
• Only 56 pts (target accrual 475)

NEJM 1996
JCO 2008
48
Effect of preoperative concurrent
chemoradiotherapy on survival of patients with
resectable esophageal or esophagogastric junction
cancer Results from a multicenter randomized
phase III study.
A. van der Gaast, P. van Hagen, M. Hulshof, M.I.
van Berge Henegouwen, G.A. Nieuwenhuijzen, J.T.
Plukker, J.J. Bonenkamp, E.W. Steyerberg, H.W.
Tilanus. CROSS study group
49
Chemoradiotherapy treatment regimen

• Chemoradiotherapy regimen
• Paclitaxel 50mg/m2 Carboplatin AUC2 on days
  1, 8, 15, 22 and 29
• Concurrent radiotherapy of 41.4 Gy in 23
  fractions of 1.8 Gy
• Surgery within 6 weeks after completion of
  chemoradiotherapy

CROSS study
50
Overall survival
188 pts surg arm 175 pts CRT arm Median F/U 32
mo
3 yr OS 59 CRT (vs 48 surg alone) p0.011
CRTx
Surgery
HR 0.67 95 CI (.49 - .91) P0.012
HR 0.67 95 CI (0.49 - 0.91)
Nos at risk
188 131 71 44 22 1
175 144 85 55 30 2
Surgery alone CRT surgery
CROSS study
51
HRs (95 CI) for death according to baseline
variables
0.67 (0.49 0.91) 0.49 (0.27 0.90) 0.72 (0.50
1.04) 0.62 (0.44 0.87) 0.92 (0.45
1.89) 0.82 (0.58 1.16) 0.34 (0.17 0.68) 0.67
(0.49 0.94) 0.67 (0.32 1.41)
Favors preoperative CRT Favors surgery alone
CROSS study
52
CROSS interesting, but questions

• Subgroup analysis no significant survival
  benefit for adenoca (74 of pts!)
• Low radiation dose is it the drugs?

53
Carboplatin-Paclitaxel as Neoadjuvant Therapy for
Esophageal Ca

• Keresztes et al. JTCVS 2003
• Carbo AUC 6, Taxol 200 mg/m2 q3wk x 2 cycles ?
  surgery
• 26 pts 100 completion full course
• - 12 grade III/IV leucopenia
• -95 improvement
  dysphagia w/in 1 wk
• -61 major clinical response, 11
  pathCR
• -3 yr OS 64 for resected patients

• DAddario et al. Onkol 2002
• Carbo AUC 3, Taxol 75 mg/m2 days 1, 8, 15
  q4wks x 2 ? surgery
• 19 pts -15.2 grade III/IV leucopenia, 3.2
  grade III/IV thrombocytopenia
• - 83 overall RR, 17 pathCR
• - 70 RR adenoca, 87 RR SCC
  esophagus
• - median F/U 19 mo 11/19 pts
  alive

54
Neoadjuvant Chemo RCTs
Neoadjuvant Chemorad RCTs
?
MRC OEO2
Walsh
?
MAGIC
Tepper CALGB
?
½?
FFCD
CROSS
?
?
55
Need a head-to-head comparison!
56
Neoadjuv CT Neoadjuv CRT p
No. of pts 58 64 -
Postop mortality () 0 6 0.12
postop complications 33 48 0.09
Recurrence rate 33 28 0.43
PathCR 3 11 0.02
5 yr DFS 21 31 0.68
Retrospective single institution cohort
57
RCT neoadjuvant chemoradiation vs. chemo for
esophageal ca

• Stahl et al. J Clin Onc 2009
• N119, adenoca distal esophagus, GEJ
• Randomized to
• Induction chemo 2.5 cycles cis, 5FU, leucovorin
• Induction chemorad 2 cycles a) 30 Gy
  concurrent cis, etoposide

58
RCT neoadjuvant chemoradiation vs. chemo
inconclusive

• No difference b/w grps R0 resection rate
• CRT grp - significant increase
• pathCR (15.6 vs. 2)
• Non-significant trend toward improved survival
  CRT grp (3 yr OS 47.4 vs. 27.7, p0.07)
• CRT 3-fold increase post-op mortality
  (10.2 vs. 3.8, p0.26)
• BUT -underpowered, closed early, non-standard
  chemo used

Stahl et al. J Clin Onc 2009
59
Summary

• Better local control
• Improved pathologic CR rate
• No added morbidity / mortality with addition of
  XRT
• Improved survival evidence from RCTs
• there is currently no conclusive evidence that
  neoadjuv CRT superior to neoadjuv CT for
  resectable esophageal / GEJ ca
• optimal multimodality tx for locally advanced
  esoph ca unknown
• a proper RCT is needed to answer this question!

60
Current feasibility trial in Halifax
Neoadjuvant chemoradiation -5 cycles carboplatin
50 mg / m2 IV paclitaxel AUC2 days 1, 8, 15,
22, 29 -45- 50.4 Gy in 25-28 fractions of 1.8
Gy / fraction, 5 fractions / wk
R A N D O M I Z E
Eligibility -age lt75 -cT2-3Nx, T1N1, M0
esophageal / GEJ adenoca or SCC -gt20 cm from
incisors, lt2 cm cardia (EGD, CT/PET) -ECOG
0-2 -no previous ca last 5 yrs,
no major comorbidities
S U R G E R Y
RESTAGING CT / PET if no mets, resectable
Neoadjuvant chemotherapy -6 cycles carboplatin 75
mg / m2 IV paclitaxel AUC3 days 1, 8, 15
q4wks x 2
61
Feasibility trial - status

• Local grant funded
• Health Canada approved Aug 3, 2011 CANADA
  PROTOCOL RECORD 147937-7
• Trial registered Clinicaltrials.gov
  Identifier NCT01404156
• Plan to begin patient enrollment Oct-Nov 2011
  pending final REB approval
• Study introduced NCIC GI oncology grp 2011
  Spring mtgformal proposal 2012
• Stay tuned!

62
More better??
63
Practice Patterns in the Treatment of Resectable
Esophageal Carcinoma A National Survey of
Canadian Thoracic Surgeons

• Emails with online survey link (SelectSurvey TM)
  were sent to all 104 members of the Canadian
  Association of Thoracic Surgeons
• Open Feb 15 July 15 2011
• 2 reminder emails sent to increase participation
  rate

64
Practice Patterns in the Treatment of Resectable
Esophageal Carcinoma A National Survey of
Canadian Thoracic Surgeons

• Response rate 54 (56 out of 104)
• 85 exclusively practiced general thoracic
  surgery
• 87 - University-affiliated hospital
• 55 - gt9 esophagectomies for esophageal cancer in
  the past 12 months

65
Number of years in practice


66
At my institution this patient would be treated
with
A 55 year old male with dysphagia and weight
loss is diagnosed with a clinical stage T3N1M0
adenocarcinoma of the distal esophagus, not
involving the gastric cardia on endoscopic
examination.  The tumor appears resectable on CT
and PET, and the patient is otherwise healthy. 
67
According to the current available evidence, this
patient would BEST be treated with
A 55 year old male with dysphagia and weight
loss is diagnosed with a clinical stage T3N1M0
adenocarcinoma of the distal esophagus, not
involving the gastric cardia on endoscopic
examination.  The tumor appears resectable on CT
and PET, and the patient is otherwise healthy. 
68
Neoadjuvant chemotherapy provides equivalent
survival benefit to neoadjuvant chemoradiation


69
For neoadjuvant chemotherapy and chemoradiation
for locally advanced, resectable carcinoma of the
esophagus, there is insufficient evidence to
conclude that one approach is superior to the
other


70
I would support a multi-center randomized trial
of neoadjuvant chemotherapy vs. chemoradiation
for locally advanced resectable esophageal
carcinoma and enroll patients if such a trial
were active