Early lung cancer: Surgery & Radiosurgery have similar results

1. Lung cancer
• Early lung cancer: Surgery & Radiosurgery have similar results
• However, NO level I evidence regarding role of Surgery or SBRT in early NSCLC
• In meta-analysis, SBRT is similar to surgery after match pair analysis
• RT dose should be BED10 >100 Gy
• 54-60Gy/3# is preferred treatment fractionation schedule
• Fiducial based Robotic Radiosurgery (CyberKnife) have not ITV & irradiated
volume is lowest
• Prospective ongoing studies will provide answers

2. Carcinoma lung: Management option
Majority (80%) of NSCLC present with advanced disease (stage III & IV)
2-Yr Survival in advanced NSCLC is ONLY 30-40%
ONLY 20% of NSCLC present at early stage (Stage I&II)
Stage I NSCLC: treatment is surgery & 5 Yr-OS is 70-80%
However, a proportion of patient with poor lung function or PS are not suitable for
urgery and are candidate for alternative treatment, eg: Conf RT, Cryotherapy,
---therapy & SBRT
So, SBRT in primary lung cancer is suitable in only a very small cohort of patient
With modern SBRT techniques, RT has become more relevant option in early lung
But, surgery has also become more effective and less risky with VATS

3. Stage I - III: Outcome after Surgery
Fry WA, Cancer 1999

4. NSCLC: SURGERY is the best option
• Morbidity and mortality in elderly and in patients with co-morbidities
• Requirement for extensive resection in 10-20%
• Deterioration in QOL
• Relapses 30%, second tumors 1-2% patients per year
• VATS is equally effective
• Overall surgical mortality-5.2 %
• Guidelines by the British Thoracic Society “ Surgery-related mortality considered
acceptable if < 4 % for lobectomy , < 8 % for pneumonectomy”
Conformal RT dose: 60-66 Gy/30-33#
Survival function: 2-Yr OS is 60-70%
5-Yr OS is 20-30%
Lung toxicity: Symptomatic pneumonitis in
15-25%
Lung toxicity depends upon mean lung
dose & V20, V10
Oesophageal toxicity 5%

5. SCRT for operable NSCLC
Onishi IJROBP 2011

6. Crabtree et al, J Thorac Cardiovasc Surg 2010
Surgery Vs Radiosurgery: Ph II study
Surgery, n=462; Radiosurgery, n=76
Overall Survival: Surgery better than Radiosurgery

7. Surgery Vs Radiosurgery: Ph II study
Crabtree et al, J Thorac Cardiovasc Surg 2010
Disease free Survival: Surgery similar to Radiosurgery

8. Crabtree et al, J Thorac Cardiovasc Surg 2010
Surgery Vs Radiosurgery: Ph II study
Propensity matched analysis
OS, DFS: Surgery= Radiosurgery

9. Early lung cancer: Surgery Vs Radiosurgery: Meta-analysis (n=864)
Meta-analysis of matched pair analysis
Jhang B et al, Radiat Oncol 2014

10. Early lung cancer: Surgery Vs Radiosurgery: OVERALL SURVIVAL
Meta-analysis of matched pair analysis
SURGERY: BETTER

11. Early lung cancer: Surgery Vs Radiosurgery: DFS
Meta-analysis of matched pair analysis
DFS SIMILAR

12. Early lung cancer: Surgery Vs Radiosurgery: Distant Control
Meta-analysis of matched pair analysis
DC: SBRT BETTER

13. Early lung cancer: Surgery Vs Radiosurgery: LOCAL CONTROL
Meta-analysis of matched pair analysis
LC: SBRT BETTER

14. Improving efficacy of RT in early NSCLC
Definite dose response relation- higher the dose higher the control
SBRT allows dose escalation
Higher dose/Fr allows more BED
BED>100 Gy- local control 84%, BED<100Gy- LC- 37%.
Majority of hypofractionation schedules are 3 fractions of 15-20Gy/Fr

15. Dose Prescription

16. All doses were prescribed
at the PTV encompassing
80% isodose.
T1 tumor- 20Gy(18)X3
Without extensive
contact with chest wall
and mediastinum
180Gy
T2 tumor and T1 with
broad contact with thoracic
wall- 12Gy(11)X5
132Gy
Tumors close to heart,
hilus or mediastinum
7.5GyX8
105Gy
421 patients
2 yr. OS -70%
3 yr. LC-97%

17. Dose/Volume constraints
Kong et al, IJROBP, 2010

18. Optimum dose for SBRT in early NSCLC
•Meta-analysis of Thirty-four observational studies with a total of 2,587 pts
•BED divided into four dose groups:
A) low (<83.2 Gy)
B) medium (83.2–106 Gy)
C) medium to high (106–146 Gy)
D) high (>146 Gy)
•The OS for the medium or medium to high BED (range, 83.2–146 Gy) was
higher than those for the low or high BED group
•BED10 need to be between 100-150Gy
Zhang, IJROBP 2011

19. SBRT: Studies 1
Munshi et al, Radiat Oncol 2014

23. SBRT in early lung cancer studies
• There are many single institution studies
• No multicentric study available
• No prospective comparison between Conv RT Vs SBRT
• No prospective study between Surgery Vs SBRT
SBRT studies:
• 2 Yr OS 70-80%
• 5 Yr OS 20-30%
Toxicity:
Gr-3/4 Pnuemonitis <5%
Rib fracture/ plexopathy <1%

24. Timmerman et al JCO 2006
Central lung tumours: Poor prognosis
• Poor prognosis of central lung tumours are mostly because of critical structures
• Lower total RT dose (BED)
• Lower dose per fraction
• Poor contouring Milano et al , Radiat Oncol , 2009

25. Failure pattern after SBRT
• 91 patients (Washington Univ)
• Most had comorbidities (poor PS or LF)
• 83 peripheral (18GyX3), 8 central (9Gy X5)
• Median FU 18 m, 2 yr LC 86 %
• 45% failures are distant ONLY
• ONLY 3% had local ONLY failure
Failures
Bradley IJROBP 2010

26. Toxicity of SBRT (n=206)
Chest wall pain 11 (5%)
Radiation pneumonitis 7 (4%)
Rib fracture 4 (2%)
Pleural effusion 4 (2%)
Atelectasis 1 (1%)
No toxicity 51%
Fatigue 28%
Chest wall pain 12%
Nausea 9%
Dyspnea 6%
Cough 6%
Erythema 2%
Hemoptysis 1%
Palpitations 1%
(Sub)acute
Late
≥ Grade 3 Radiation Pneumonitis 0-5 %
Radiation induced plexopathy (upper lobe tumours)
Late Chest wall toxicity – Up to 10 %
Lagerwaard FJ, IJROBP, 2008
Petterson et al, Radiother Oncol 2009

27. Ong CL et al, Radiat Oncol 2010
Lung toxicity: Dose-effect relationship
Symptomatic pneumonitis depends upon V20, V15
Asymptomatic pneumonitis depends upon low dose volume

28. Yamashita H et al, WJR 2014
Lung toxicity: Dose-effect relationship
Radiation pneumonitis Gr 3-4: 2-21%
Depends upon mean lung dose, V20 & damaged lung vol dose

29. Jin JY IJROBP 2009
Lung toxicity depends upon relative lung damage vol
Damaged lung is the volume receiving
threshold dose of RT
High dose SBRT: higher dose per fraction
will deliver less dose to lung
Low dose RT: lower dose per fraction
Small volume peripheral tumour need high
dose per fraction to increase efficacy (high
BED; >100Gy) & low damaged lung volume
Hence high local control, low toxicity

30. Amini et al. Radiation Oncology 2014
Re-irridiation: Dose-effect relationship

31. Hypo fractionated SRT in lung cancer
Likely candidates:
•Small lesions (< 5 cm)
•Histologically proven (especially in India!)
•Poor performance status
•No nodes/distant metastasis
Medically Inoperable patients:
•Compromised Pulmonary Function
– Pulmonary spirometry, Arterial Blood gases
– TLCO, maxi O2 uptake
– Prior major resection as pneumonectomy
– Ventialtion/Perfusion scan
•Ischemic heart disease
•Left ventricular dysfunction (2 D Echo)
•Pulmonary vascular diseases
– Precapillary PAH
– Postcapillary PAH

32. Robotic Radiosurgery
Pencil beam: small lesions
Melanoma eye, Trigiminal
Multiple isocentre: Irregular tumour
Skull tracking: brain tumours
Fiducial based tracking: prostate
Fiducial tracking with syncrony: lung cancer
Real time tracking for fiducial
NO ITV margin
Treated lung volume low

33. Treatment Margins
GTV
PTV
Conventional SBRT SBRT
ITV

34. Treatment Margins
GTV
PTV
Conventional SBRT Fiducial based Robotic
Radiosurgery
ITV
CTV
NO ITV for fiduicial based real time tracking

35. • Stage I NSCLC treated with SBRT between 2004 and 2011 in
National Cancer Database
• Overall mean BED10 = 134.5 and median BED10 =132 Gy
• 94.5% were prescribed a regimen with a BED10≥100 Gy
•
• Most common prescriptions: 60 Gy/3# = 24%, 48 Gy /4#
=17.8%; 50 Gy/5# = 13.0%; 54 Gy/3# =12.8%
• Decreased utilization of 54 to 60 Gy in 3 fractions (47.9% in
2006 to 27.9% in 2011, combined) and increased utilization of
50 Gy in 5 fractions (3.1% in 2006 to 20.4% in 2011).
• Majority of patients being treated with regimens employing a
BED10≥100 Gy.
• Since 2006, decline in the use of 54- 60 Gy /3#, increase in
use of 50 Gy/5#
• Possible explanations: 1) concern for increased toxicity with
higher BED regimens, 2) increasing treatment of centrally
located tumors.
Corso CD et al, Am J Clin Oncol 2014
Stage I Lung SBRT Clinical Practice Patterns (n=5246)

36. • National Cancer Database from 2003-2011
• T1-T2N0M0 inoperable lung cancer (n= 39,822)
• Logistic regressions were performed to determine
predictors of receiving any 1) radiation vs. No Tx
and 2) receiving SBRT vs. ConvRT.
• Treatment with RT significantly less likely in blacks
(OR 0.65) and Hispanics (OR 0.42) compared to
whites.
• Treatment with SBRT vs ConvRT was more likely in
an academic research program (OR: 2.62) and a
high-volume facility (OR: 7.00) compared to
community cancer programs or low-volume
facilities.
Koshy M et al; J Thoracic Oncol 2014
Disparities in Treatment of Patients with Inoperable Stage I Non-Small
Cell Lung Cancer: A Population-Based Analysis (n=39822)
Marked institutional and socioeconomic variations in the treatment of inoperable stage I NSCLC
Yr-2011

37. Surgery Vs Radiosurgery studies:
-ROSEL study
-STARS study
-JCOG 0403 study
Radiosurgery Vs Conv RT studies:
-RTOG 0915
Arm 1: 37Gy/1#
Arm 2: 48Gy/4#
-RTOG 0813
Dose escalation study
-TROG 09.02
Arm 1: 66Gy/33#
Arm 2: 54Gy/3#
-SPACE
Arm 1: 70Gy/35#
Arm 2: 45Gy/3#
On going studies in SBRT lung

38. Lung cancer
• Early lung cancer: Surgery & Radiosurgery have similar results
• However, NO level I evidence regarding role of Surgery or SBRT in early NSCLC
• In meta-analysis, SBRT is similar to surgery after match pair analysis
• RT dose should be BED10 >100 Gy
• 54-60Gy/3# is preferred treatment fractionation schedule
• Fiducial based Robotic Radiosurgery (CyberKnife) have not ITV & irradiated
volume is lowest
• Prospective ongoing studies will provide answers