2018 preceptorship-lung-stereotactic-ablative-rt-early-stages-nsclc-suresh-senan

Stereotactic ablative radiotherapy (SBRT / SABR)
for early-stage lung cancer
Professor Suresh Senan, VU University Medical Center

Disclosures
Research grants: Varian Medical Systems, ViewRay Inc.
Advisory boards – Eli Lilly, AstraZeneca, Merck

ESMO Clinical Practice Guidelines
(Postmus PE, 2017)
• Eligible patients
• Toxicity in peripheral tumors
• Post-treatment follow-up
• Higher risk tumors [central, ILD]
Outline of talk
Stereotactic ablative radiotherapy (SABR, SBRT)
SABR is a technique for delivering external
beam radiotherapy to an extra-cranial target
• with a high degree of accuracy
• using high doses of irradiation
• 1-8 treatment fractions
IASLC Multidisciplinary Approach to Thoracic
Oncology (2014)

Imaging and delivery of SABR (typical examples)
4-Dimensional CT scan
Cone-beam CT scan Volumetric Modulated Arc
Therapy (FFF delivery in <4 mins)
Linear accelerator
Doses should be ≧BED10100 Gy
[Vansteenkiste J, ESMO guidelines 2014]
ESTRO ACROP consensus guidelines on implementation and
practice of stereotactic body radiotherapy for peripherally located
early stage NSCLC [Guckenberger M, Radioth Oncol 2017]

‘Risk-adapted’ SABR fractionation schemes
3 fractions of 18 Gy: T1 lesions, not adjacent to chest wall
5 fractions of 11 Gy: Broad chest wall contact, T2 lesions
8 fractions of 7.5 Gy: Central lung lesion
Hurkmans C, Rad Onc 2009
Single-fraction SABR (prospective clinical trials)
RTOG 0915, Videtic GM, IJROBP 2015: 34 Gy
Multi-institutional, Singh A, IJROBP 2017: 30 Gy
Yamamoto N, JTO 2016: ≧36 Gy (RBE, carbon ions)

Recurrences Local Regional Distant
MD Anderson Hospital1
912 patients; median follow-up
of 59 months
11% 12% 21%
VU University Med Center2
676 patients; median follow-up
of 33 months
10.5% 12.7% 20%
SABR: Recurrence patterns in stage I NSCLC
1 Brooks E, IJROBP 2017 and update at ASCO (Abstr 8501)
2 Senthi S, Lancet Oncol 2012

SABR toxicity in peripheral tumors
• No decrease in quality of life - systematic review of 9 studies 1
• Chest wall pain, grade 3, or higher in between 1-19% of patients 2
• Radiation pneumonitis rates in 504 SABR patients:3
• Grade ≥ 3 pneumonitis in 2% of patients without interstitial lung disease (ILD)
• Grade ≥ 3 pneumonitis in 32% of patients with ILD
• Grade 5 pneumonitis in 21% of patients with ILD
1 Chen H, Clin Lung Cancer 2016; 2 Shaik T, Cancer Treat Rev 2014; 3 Bahig H, Prac Radiat Oncol 2016

SABR versus conventional radiotherapy
SPACE trial [Nyman J, Radioth Oncol 2016]
102 stage I patients randomized between 2007-201
Primary endpoint: progression free survival at 3 years
Local control similar: SABR - 86% vs conventional - 86%
SABR had superior HRQL evaluation (EORTC QLQ 30, LC14 modules): After 3DCRT, worse
dyspnea (p = 0.01), chest pain (p = 0.02) and cough (>10 points difference) observed
CHISEL trial [Ball D, WCLC 2017]
101 stage I patients randomized between 2009-2015
Primary endpoint: time to local failure
With SABR,
SABR led to superior freedom from local failure (HR = 0.29, 95% CI 0.130, 0.662, P=0.002)
SABR resulted in longer overall survival (HR = 0.51, 95% CI 0.51, 0.911, P=0.020)

SABR: Challenges in practice
Establishing a tumor diagnosis in less fit patients
Response assessment and salvage therapies (multi-disciplinary)
Higher-risk populations for SABR
Lack of reliable organ at risk tolerance data

Proceed with treatment Treatment discouraged
ESMO 2014 Yes, if MDT agrees
ASTRO 2017 Yes, if MDT agrees In some regions
Asian Guidelines 2016 No Early biopsy preferred
Guidelines on treatment without pathology
Vansteenkiste J, Ann Oncol 2014; Videtic GMM, Pract Rad Oncol 2017; Bai C, Chest 2016
MDT = multi-disciplinary team

Diagnosis of clinical stage I NSCLC by Dutch MDT’s
Dutch Lung Surgery Audit: 1555 clinical stage I NSCLC cases – a final diagnosis of
benign disease was made in 0.8% [Heineman D, ATS 2016]
Intraoperative diagnosis (IOD) in Dutch Cancer registry [Verhaegen A, 2017]
IOD rates of 55%, 39% and 23% after segmentectomy, lobectomy and bilobectomy,
respectively
IOD rates highest in small tumors (cT1A - 59%)
IOD rates varied between hospitals (15% - 65%)
http://eacts2017.event.y-congress.com/ScientificProcess/data/22/64/302/ff0df5b9-708e-4739-b408-0bc30f423c3e/Uploads/000203.pdf

Serial images in a patient without recurrence
Ronden M, IJROBP 2017
High-Risk Features (HRF)
Enlarging Opacity
Sequential Enlargement
Enlargement after 12 months
Bulging Margin
Linear Margin Disappearance
Loss of Air Bronchogram
Cranio-Caudal Growth
Huang K, Radioth Oncol 2014
Post-SABR radiological changes

Ronden M, Pictorial atlas. JTO 2018

International Delphi consensus [Ngyuen T, Pract Rad Oncol 2017]
Suspicious of a local recurrence are: infiltration into adjacent structures, sustained growth, bulging
margins, spherical growth, mass-like growth, cranio-caudal growth and a loss of air bronchograms
FDG-PET/CT scans recommended only when there was suspicion for a local recurrence
Ronden M, JTO 2018

Post-SABR evaluation
Due to a high number of false-positive
findings on PET, patients suitable for
salvage therapy should undergo a
biopsy, whenever possible [III, B]
ESMO Clinical Practice Guidelines
Postmus PE, Ann Oncol 2017
Huang K, Radioth Oncol 2013

Post-SABR surgical salvage appears safe
• Chen F, J Thoracic Oncology, 2010
• Neri S, J Thoracic Oncology, 2010
• Hamamoto Y, Japan J Radiology 2012
• Allibhai Z, Eur Resp Journal, 2012
• Hamaji M, J Thoracic Oncology, 2015
• Verstegen N, Radioth Oncol 2016
• Antonoff MB, JTCVS 2017

Stage I NSCLC: New primary tumors
855 post-SABR patients from VUMC
[Verstegen NE, JTO 2015]
1294 surgical cases from MSKCC
[Lou F, JTCVS 2012]
SEER database [Thakur MK, JTO 2018]: Rate of SPLC development is 1.10% per patient per year,
with median time intervals between the IPLC and SPLC diagnoses of 59 and 62 months

Moderately central tumors Ultracentral tumors
SABR is relatively safe
ASTRO guidelines [Videtic
GMM, Prac Rad Onc 2017]
Toxicity with high-dose radiation
[Haseltine JM, PRO 2016, Tekatli H,
JTO 2016; Lindberg K, WCLC 2016]
Central and ‘ultracentral’ tumors

Radiographic and clinical toxicity
Main stem bronchi (n=9), intermediate bronchus (n=8),
upper lobe bronchi (n=25), middle lobe bronchi (n=6),
and lower lobe bronchi (n=33)
195 central tumors (585 bronchial structures)
treated using ≤12 fractions between 2006-2015,
at 2 Dutch centers
Doses recalculated to an equivalent dose of 2Gy
with an α/β ratio of 3
Radiographic toxicity defined as:
• Airway stenosis
• Occlusion without atelectasis
• Occlusion with atelectasis
Tekatli, H, IJROBP 2017

Images of a patient treated with 12 fractions of 5 Gy. CT scans 12.7 months later show stenosis
of the left main bronchus PLUS occlusion with atelectasis of the left lower lobe bronchus
Pooled analysis: Radiological toxicity

• Radiographic bronchial toxicity in 28% of patients
• Clinical ≥G3 toxicity in 12%
• Predictors for ≥G3 toxicity on multivariate analyses
(i) Tumor volume (PTV) overlapping trachea / main stem bronchus (p=0.005)
(ii) COPD (p=0.034)
(iii) total V130Gy,EQD (p=0.012)
Radiographic and clinical toxicity

• Moderately central tumors [G5 toxicity possible or likely in 7.5% of patients,
Tekatli H, Radioth Oncol 2015]
• Interstitial lung disease [G5 toxicity in 15%, Chen H, IJROBP 2017]
• Re-irradiation [G3 or higher toxicity in 5-30%, De Bari B, Ca Treat Rev 2015]
• Paramediastinal tumors, and lesions close to left hemi-diaphragm
SABR in higher-risk populations

MRI-guided adaptive SABR for high-risk SABR

Senan S, Proc ASTRO 2017
MRI-guided SABR during breath-hold

Guidelines in operable stage I NSCLC
ASTRO Guidelines [Videtic GM, Pract Rad Oncol 2017]:
For patients with “standard operative risk” (i.e. with anticipated operative mortality
of <1.5%) and stage I NSCLC, SABR is not recommended as an alternative to
surgery outside of a clinical trial
Endorsed by European Society for Radiotherapy & Oncology (ESTRO) and the IASLC
ESMO Guidelines [Vansteenkiste J, Ann Oncol 2014]:
Surgery should be offered to patients who are willing to accept procedure-related risks
SABR is the preferred treatment in patients with a peripheral early-stage NSCLC who
are unfit for surgery, or who refuse it

4 previous unsuccessful attempts at performing prospective randomized trial
ROSEL NCT00687986
STARS NCT00840749
ACOSOG-4099/RTOG1021 NCT01336894
Mayo Clinic NCT01622621
Randomized trials of surgery versus SABR
Lancet Oncol 2015

Local treatment for peripheral lung tumors
Table 1: 5-year survival rates after SABR in operable patients of 51-74%

Non-randomized comparisons of surgery vs SABR
National Cancer Database (NCDB): overall survival for stage IA NSCLC wedge resection vs SABR
Wedge excision
SABR
Yerokun BA, JTCVS 2017

Pitfalls of using NCDB for outcomes research
Post-surgical OS in 972 stage I patients operated at Duke Univ,
stratified by predicted DLCO [Berry MF, ATS 2015]
Population study on COPD outcomes. [Lange P, AJRCCM 2012]
National Cancer Database does not capture data on
pulmonary function, cardiac function, performance
status.
Data on cancer-specific or disease-free survival is
unavailable [Rosen JE, Ann Thorac Surg 2014]

Chen H, IJROBP 2017
In the absence of completed clinical trials, propensity score studies have attempted
to mimic clinical trials using non-randomized data.
We identified 16 studies comparing SABR and surgery
• In patients eligible for either treatment, better overall survivals were seen after surgery compared to SABR
• Similar lung cancer-specific survival for both treatments
• Propensity score-matching caliper distance and first author specialty were associated with survival
endpoints on regression
Conclusions: Prospective clinical trials are preferred to propensity analyses in
evaluating the nature of non-cancer related mortality post-SABR
Meta-analysis of propensity score studies

Requirements of new studies in early-stage NSCLC
• Cause-specific outcome analysis
• Morbidity and mortality analysis for at least 1 year post-treatment
• Subgroup analysis for age,Charlson comorbidity index, and histologic subtypes
• Long-term pulmonary and cardiac function
Only such data can justify the rationalization to choose optimal curative-intent
therapy for patients with stage I NSCLC.
T Eguchi, JCO 2017

Role of shared decision making (SDM)
UK Supreme Court ruling in 2015
Dutch SDM website
http://www.keuzehulp-longkanker.nl/
SDM is a process in which patients and physicians discuss the current
evidence on different treatment options and mutually arrive at a clinical
management plan [Stacey D, CA Cancer J Clin 2008]

Developments in the United Kingdom

Early-stage lung cancer: A preference sensitive disease?
Preference-sensitive care describes a situation where the
evidence for the superiority of one treatment over another
is not available; there are therefore two or more valid
approaches to care and the best choice depends on how
a patient values the risks and benefits of the treatments
available.
Elwyn G et al, Implement Sci 2009. Dual equipoise shared decision
making: definitions for decision and behaviour support interventions
Readmission rates
England - Moller H, Eur J Cancer 2016
US - Stiles BM, ATS 2016
90-day mortality after surgery and SABR
NCDB – Stokes WA, JCO 2018
6 month mortality after adjuvant chemotherapy
NCDB - Morgensztern D, JTO 2018
Survival from competing causes of death
Eguchi T, JCO 2017
Soneji S, Chest 2017
ESMO Guidelines [Vansteenkiste J, Ann Oncol 2014]:
SABR is the preferred treatment in patients with a peripheral early-stage NSCLC who are unfit for surgery, or who refuse it

90-day mortality for early-stage NSCLC (NCDB)
Stokes WA, JCO 2018
90-day mortality in Europe
English, lobectomy1
70-80 years PS 0: 4%
70-80 years PS 1: 6%
Dutch, lobectomy2
70-79 years: 4.4%
1 O’Dowd EL, Lung Cancer 2016
2 iknl.nl

Trends in early-stage NSCLC [Netherlands]
Netherlands Cancer Registry (n = 21,032 patients)
[Louie AV, Lung Cancer 2016]
Diagnosis year
1997-1999
Diagnosis year
2009-2011
Surgery 62% 60%
Radiation 19% 28%
Palliative therapy 19% 13%
49% of Dutch pulmonary physicians consider surgery and SABR
to be equal treatment options [Hopmans W, Radioth Oncol 2015]
71% of all lung cancer resections are performed by VATS
[Detillon DE, ATS 2017]

Take-home messages
SABR is the preferred treatment in patients with a peripheral early-stage NSCLC
who are unfit for surgery, or who refuse it [ESMO Guidelines 2017]
Early-stage lung cancer is a preference-sensitive disease
Growing data on SABR outcomes in patients fit to undergo surgery [Weder W,
EJCTS 2018]. Results of prospective, randomized clinical trials are awaited
SABR-related toxicity is higher in patients who have centrally located tumors, and
those with pre-existing interstitial lung disease

Re-admission after surgery and discharge status
• English National cohort analysis of 15,738 resected NSCLC patients in 2006-2010
[Moller H, Eur J Cancer 2016]
• 20% and 45% were re-admitted as in-patients within 30 and 90 days, respectively
• State Inpatient Databases (Healthcare Cost and Utilization Project); all lobectomies
from 2009 to 2011 in California, Florida, and New York [[Moller H, Eur J Cancer 2016]]
• OL = open lobectomy; MIL = minimally invasive lobectomy
• 22,647 lobectomies identified (58.8% OL vs 41.2% MIL; median age, 68 years; median
length of stay, 6 days)
• Most patients (59.8%) had routine discharge home
• Other home health care, 29.4%; transfer to other facility, 8.8%; mortality, 1.9%.
• 90-day readmission rate was 19.8% (OL 21.1% vs MIL 17.9%, p < 0.001).

Competing risk analysis in NSCLC survivors
Resected stage I NSCLC at
MSKCC [Eguchi T, JCO 2017]
Deaths after local treatment for early-stage NSCLC
≧75 years
CumulativeIncidenceofDeath
In patients ≧ 65 years of age, noncancer-
specific CID was higher than lung cancer–
specific CID for up to 2.5 years after resection
Higher noncancer-specific, early-phase
mortality was enhanced in patients ≧ 75 years
of age than in those 65-74 years of age
Multivariable analysis: the predicted postoperative
(ppo) diffusing capacity of the lung for carbon
monoxide a significant predictor of 1 year mortality,
and noncancer specific deaths

Journal of Thoracic Oncology DOI: (10.1016/j.jtho.2018.01.010)
Early Mortality after adjuvant chemotherapy for NSCLC
Morgensztern D, Journal of Thoracic Oncology 2018
NCDB analysis of mortality within 6 months in 19,398 patients with Stage IB to
IIIA NSCLC (2004-2012), and who received multiagent adjuvant chemotherapy
starting within 120 days from resection with negative surgical margins

2018 preceptorship-lung-stereotactic-ablative-rt-early-stages-nsclc-suresh-senan

VALOR (USA) POSTILV (China) SABRTooth (UK) STABLE-MATES (USA)
Eligibility
criteria
Tumor ≤5cm
(peripheral
and central)
Tumor ≤3 cm, fit
for lobectomy or
pneumonectomy
High-risk operable,
peripheral
tumors ≤5cm,
High-risk operable,
patients pre-
randomized
Primary End-
point
5-year overall
survival
2-year local-
regional control
Average recruitment
rate of 3 pts/month
for a 15 month
period
3-year overall survival
Secondary
end-points
QoL, patterns
of failure,
cause of death
OS, DFS, site-
specific failure,
Time to LR failure
and DM
PFS, failure patterns,
toxicity, and 5-year
overall survival
Planned
accrual
670 76 54 (feasibilty phase) 258
Randomized trials of surgery versus SABR

2018 preceptorship-lung-stereotactic-ablative-rt-early-stages-nsclc-suresh-senan

More Related Content

What's hot (20)

Similar to 2018 preceptorship-lung-stereotactic-ablative-rt-early-stages-nsclc-suresh-senan (20)

Recently uploaded (20)

2018 preceptorship-lung-stereotactic-ablative-rt-early-stages-nsclc-suresh-senan