Respiratory non-small cell lung cancer definitive stereotactic EBRT central tumours

• Non small cell lung cancer (NSCLC).
• T1-2 (≤5 cm maximum diameter) N0 M0.
• Centrally located tumour defined as ≤2 cm around the proximal bronchial tree (PBT).^r  ​
  • PBT is defined as the distal 2 cm of trachea, carina, and named major lobar bronchi up to their first bifurcation.^r
  • This protocol does not apply to ultra-central tumours. Ultra-central defined as having a planning target volume (PTV) touching or overlapping the central bronchial tree, oesophagus, pulmonary vein, or pulmonary artery.^r
• Medically inoperable or declining surgery.
• Preferably histologically confirmed NSCLC (alternatively, high pre-test probability for malignancy as determined by consensus at a thoracic MDT).

• ECOG 0-2.

Cautions

• Interstitial lung disease.

• Previous high dose radiation therapy to the chest region to be treated.^r

• Cardiac implantable electronic devices (CIEDs).

• Non-rheumatoid collagen vascular disorders.

Exclusions

• Pregnancy.

Indications notes

• This protocol does not apply to peripheral tumours. Link to Respiratory non-small cell lung cancer definitive stereotactic EBRT peripheral tumours.

• Endobronchial ultrasound and biopsy of enlarged or PET positive nodes may be considered to confirm staging.

Tumour motion

• Tumour motion must be assessed in 4D to capture motion in each direction.
• Consider if motion management is beneficial and determine the optimal motion management strategy to reduce or control respiratory motion (motion <10 mm is less likely to benefit).^r
• If appropriate, consider one of these methods: 
  • Breath hold (BH) (inspiratory or expiratory).
  • Abdominal or vacuum compression (i.e. drape).
  • Gating.
• With advanced motion management strategies, strongly consider multiple imaging studies prior to treatment commencement to test set up reliability and compliance with motion management strategy (with respect to consistency of breath holds, target and fiducial marker/surrogate motion, and target volume margins). 
• Further information on methods used in the management of respiratory motion in radiation oncology can be found in the American Association of Physicists in Medicine (AAPM) Task Group Report 76^r and the UK SABR Consortium Guidelines 2019.^r
• Cyberknife^® Robotic Radio-Surgery system uses inspiratory/expiratory 3D CT for planning (rather than 4D CT).

Patient positioning and immobilisation

Planning CT acquisition

• Planning CT scan limits should be superior to the supraclavicular fossa through to inferior to the diaphragm (ensure entire lungs are scanned).
• Slice thickness ≤3 mm (≤2 mm desirable).
• Breath hold (BH) technique:
  • For patients performing breath hold a method to confirm reproducibility should be used e.g. fluoroscopy or repeat CT.
  • Strongly consider multiple imaging studies prior to treatment commencement to test set up reliability (with respect to consistency of breath holds, target and fiducial marker/surrogate motion, and target volume margins).
• Free breathing (FB) technique:
  • 4D CT and 3D CT are required.
  • If planning based on free breathing, 4D CT used for treatment planning should be reviewed to ensure it is an accurate estimate of tumour motion.

• Contrast CT to be considered depending on tumour position.

* Use flattening filter free (FFF) beams where available. 

Target volume objectives

Margins have been agreed by the Radiation Oncology reference committee and reflect safe clinical practice.

Target volume notes

• Tumour motion with advanced motion management strategies must remain accounted for in target volume margins (i.e. IGTV or ITV).
• The Cyberknife^® Robotic Radio-Surgery system tracks and automatically adjusts for movement (using a system combining infrared tracking of external chest markers with kV imaging). Margins are typically PTV = GTV + 0.5-0.7 cm (depending on tumour tracking used).

Max = Maximum.

**PTV coverage should not be compromised to meet this constraint.

OAR notes

• A maximum dose is considered to be dose to a minimum volume of 0.03 cc.
• OARs should be contoured on the appropriate scans:
  • For BH technique: breath hold CT.
  • For FB technique: average intensity projection (AIP) of a 4D CT. 
  • For gated technique: gated AIP.

Suggested OAR contouring atlases

• Kong et al. 2011^r
• RTOG lung contouring atlas. opens in a new tab or window
• SAFRON II protocol opens in a new tab or window (supplement 1).^r
• Organ at risk delineation for radiation therapy clinical trials: Global Harmonization Group consensus guidelines.^r
• Other OAR constraint guides include the AAPM Task Group 101 report^r, HyTEC^r and the UK SABR Consortium Guidelines 2019^r and UK 2022 consensus^r.

Treatment techniques

• Treatment to be delivered using conformal/DCAT or intensity modulated techniques including IMRT, VMAT, and Cyberknife^®.
• The potential for over modulation should be considered for single fraction deliveries and when respiratory motion is >10 mm.
• The choice of which CT to use for planning depends on the motion management technique used.
• Treatment is usually planned on the average intensity projection (AIP).

Treatment planning system

• Treatment planning system should have at least a superposition/convolution type dose algorithm or Monte Carlo or Linear Boltzmann Equation dose algorithms.^r
• Dose grid resolution on the final dose calculation must be ≤2 mm.^rrr

Evaluation of plan quality

• It is important to assess the quality of plans beyond achieving acceptable target coverage and meeting OAR constraints.^r

Dose conformity

• High dose spillage: cumulative volume of all tissue outside the PTV receiving a dose of >105% of the prescribed dose should be no more than 15% of the PTV volume.^r
• Dose conformity of PTV coverage can be assessed using conformity index (CI): CI 100% values closest to 1.0 indicate better conformity of dose to the target. The CI 100% will ideally be <1.2.^r See table below from SAFRON II protocol.
• Intermediate dose spillage: The CI 50% is dependent on tumour volume, shape and proximity of OARs, refer to table below.
• A relevant protocol which could be consulted is SAFRON II. For more information see the SAFRON II protocol opens in a new tab or window (supplement 1). ^r

Conformality of prescribed dose

© BMC Cancer 2016^r

Dose inhomogeneity

• Maximum within the PTV should be between 125-143% of the PD (see target objectives).

Target verification

*The Cyberknife^® Robotic Radio-Surgery system uses repeated kV imaging to image the tumour throughout treatment.

Target verification recommendations reflect the minimum imaging required for the safe delivery of this protocol.

Target verification notes

• Volumetric image guidance is mandatory and 4D image guidance is best practice.^r 
• All pre-treatment and during treatment imaging should be appropriate to treatment technique and motion management used.

Pre-treatment imaging

• Following patient set up, each patient must have a pre-treatment 3D/4D CBCT with online correction (as appropriate to motion management and treatment technique).
• A 0 mm tolerance should be considered standard for target position.
• Attention should be given to the impact of shifts close to OARs.
• Consider 6 degrees of freedom correction in the setting of multiple targets or central lung tumours.
• Repeat pre-treatment imaging is strongly recommended when shifts are ≥3 mm in any direction.

During-treatment (intrafraction) imaging

• Intrafraction CBCT may be acquired between or during arcs or trigger imaging may be used to assess target/surrogate position during beam on time.
• Consider imaging during treatment for prolonged treatments.

Post-treatment imaging

• Post-treatment CBCT is optional.

Efficacy

There is currently limited prospective evidence evaluating SABR treatment in central lung tumours. RTOG 08-13 was a multicentre, phase I/II study designed to evaluate the maximum tolerated dose, efficacy and toxicity profile of a 5 fraction schedule.^r 120 patients with cT1-2 (≤5 cm) N0 M0, centrally located, biopsy-proven, medically inoperable non-small cell lung cancer, were recruited between 2009 and 2013 to undergo 5 fraction SABR, with doses ranging between 50 Gy and 60 Gy. Median follow up was 37.9 months. For the 8 patients treated with 10 Gy per fraction, the 2 and 3-year local control rates were 87.5% and 75%, respectively and the 3-year overall survival rate was 75%. No patients at this dose per fraction level developed severe toxicities. Higher doses per fraction were associated with improved efficacy but also increased risk of severe toxicities. In those with disease recurrence, distal relapse is more common than local or regional relapses.

Eight fraction schedules such as 8 fractions of 7.5 Gy, is also commonly used for central lung tumours, particularly in Canada, the UK and Europe. Retrospective evidence supporting use has demonstrated promising outcomes so far.^r In a phase I study, Kimura et al evaluated the tolerability of the 8 fraction schedule, beginning at 60 Gy.^r No dose limiting toxicity were encountered in all 9 patients at this level. OAR constraints were not met at 64 Gy and therefore maximum tolerated dose were determined to be 60 Gy. A systemic review suggests dose prescriptions with BED ≥100 Gy₁₀ and BED ≤210 Gy₃, such as 60 Gy in 8 fractions, result in acceptable efficacy and toxicity rates.^r Of note, 60 Gy in 8 fractions is adopted for use in the ongoing phase I SUNSET study on ultra-central tumours.^r

Toxicity

Early studies have demonstrated excessive rates of treatment related toxicities (~50%) in the treatment for central lung tumours.^r Later studies have shown improved tolerability with the use of more protracted (e.g. 4 or more fractions) or lower BED schedules. RTOG 08-13 have demonstrated that the use of the 5 fraction schedule to be reasonably safe, in keeping with findings from many previous retrospective studies. However, the risk of late severe treatment related toxicities appears to be higher than expected for peripheral tumours - with grade 5 toxicities typically occurring more than one year post completion of SABR treatment.^r

Although most SABR studies on central tumours have reported the risk for grade 3 or greater toxicities to be ≤10%, a few retrospective studies have documented higher rates of severe adverse events.^rrThe recently reported prospective, phase II HILUS study demonstrated increased rates of treatment related toxicities, especially when irradiating tumours within a 1 cm of the main bronchi and trachea.^r Findings from this study may be confounded by various factors: 1) 56 Gy in 8 fractions was prescribed to the 67% isodose line, resulting hotspots of up to 150% in the target volume, 2) variable setup and target volume expansions used between participating centres, 3) wall structure was not included for luminal organs at risk, and 4) dose to some critical organs at risk were not controlled. Risk factors for pulmonary haemorrhage (one of the most common central lung SABR related morbidities) include proximity of disease to major airways, gross endobronchial disease, excessive primary bronchial tree Dmax, and use of antiplatelet/anticoagulant/bevacizumab.^rr