COLLISION-XL trial: Unresectable colorectal liver metastases: stereotactic body radiotherapy versus microwave ablation - a phase II prospective randomized controlled trial for CRLM 3-5 cm

Colorectal carcinoma is one of the most common malignancies in the Western
world. In the course of the disease 40-60% of patients develop colorectal liver
metastases (CRLM). Without treatment, survival for these patients is cumbersome
with a median overall survival (OS) of 7.4 - 11 months. Although
chemotherapeutic regimens are slowly improving, local therapy for CRLM remains
the only option associated with a realistic chance of long-term disease control
or in selected cases even full cure. Surgical resection represents the
historical standard and treatment of first choice with 5-year OS reaching
35-60%. However only a minority of the patients (i.e. 10-20%) is eligible for
surgery due to size, number or location of the metastases, or relevant
co-morbidities. To eliminate unresectable metastases, several ablative
strategies have emerged.

Thermal ablation techniques employing radiofrequency ablation (RFA) and
microwave ablation (MWA) have slowly worked their way into common clinical
practice and international guidelines. Thermal ablation for small liver lesions
has an excellent safety profile with a low complication rate for smaller liver
tumours. However, the issue of local site recurrence after thermal ablation
has prohibited widespread adoption of the technique for resectable lesions. In
the last few years, thermal ablation techniques have substantially improved
with primary efficacy rates (complete ablation after the first procedure) for
lesions <=3cm now reaching 92-100%. These results are comparable to recurrences
after surgical resection for similar-sized lesions. The relative ease to
percutaneously re-ablate potential site recurrences, nowadays often in the
setting of a one-day admission under conscious sedation, has downgraded the
relevance of local site recurrence (LSR) with local control rates (assisted
technique effectiveness) approaching 100% for lesions <=3cm. The recently
presented long-term results from the EORTC CLOCC-trial (ASCO 2015) show a clear
survival benefit of RFA plus systemic chemotherapy over chemotherapy alone for
unresectable CRLM: 8-year OS 36% vs 8.9% (p=0.01; HR 0.58; 95%CI 0.38-0.88).
Numerous studies have demonstrated a superior safety profile in addition to
lower direct and indirect costs of thermal ablation over surgical resection.

Stereotactic body radiotherapy (SBRT) is gaining interest as a potential means
to treat CRLM. Especially for solitary or a limited number of CRLM the
potential to induce long-term local tumour control has been established with
acceptable toxicity. Several recent propensity matched comparisons from the
radiation oncology community seem to favour SBRT over thermal ablation for
lesions >3cm, substantiated by a superior long-term freedom from local
recurrence rate following the initial treatment.

In their response, the interventional oncology communities state that simply
comparing local control rates following an ablative procedure seems unjust when
comparing an easily repeatable technique with a one-shot treatment method.
Furthermore, the claim that SBRT is less-invasive because it is a no-needle
technique seems unsubstantiated and may well be prejudiced given the
unquestionably larger area of collateral damage following SBRT.

Given the results from the EORTC-CLOCC trial, the comparable survival for
ablation*+*surgery versus surgery alone, the potential to induce long-term
disease control and the very low complication rate following percutaneous
ablations for CRLM <= 3 cm and given the fact that the local control rate has
approached 100% for smaller CRLM, we believe SBRT is currently not indicated
for thermally ablatable CRLM <= 3 cm.

However, thermal ablation local control rates clearly descent with increasing
tumour-size and the number of complications will rise. Hence, we believe
clinical and oncological equipoise for CRLM 3 - 5 cm has been reached for
unresectable CRLM treated with either MWA versus SBRT.

The primary objective is to compare efficacy of MWA to the efficacy of SBRT
with regards to the primary endpoint (local tumour progression free survival at
1 year [1-year LTPFS]) in patients with unresectable CRLM (3 - 5 cm) that are
unsuitable for surgery due to either comorbidities, a history of extensive
abdominal surgery, a poor performance status or due to a certain unfavourable
anatomical location of the tumour.

COLLISION XL is a prospective multi-centre phase-II randomized controlled
trial.

SBRT or MWA. The expert panel, consisting of at least two interventional
radiologists, two hepatobiliary surgeons and two radiation oncologists, will
appoint lesions of 3-5cm, that are unresectable and suitable for both MWA and
SBRT, as target lesions.

Over the last decades, technical developments made it possible to deliver high
radiation doses per treatment fraction more precisely to the tumour, called
stereotactic body radiotherapy (SBRT). Furthermore, in some hospitals it is
possible to visualize the tumour during radiation treatment to deliver gated
treatment (beam-on only when the tumour is in the predetermined position) using
small uncertainty margins and thereby limiting the dose delivered to
surrounding normal organs, likely resulting in decreased toxicity.
Disadvantages include the need to be positioned within the MRI bore during
radiation delivery, and a prolonged time per treatment fraction. Local tumour
control of SBRT for liver malignancies ranges between 50-95% after one year. A
recent systematic review showed a one-year local control rate of 67% and a
two-year local control rate of 59.3%; however, this systematic review also
included older studies and in the last few years SBRT techniques have
substantially improved. Grade I-II toxicity occurred in 23-78% of patients
receiving SBRT, grade III toxicity or higher only occurred in 0-10% of
patients.
Reported outcomes after thermal ablation for CRLM are improving. Survival
results after thermal ablation for non-surgical candidates for lesions <3cm
have approached the results achievable with surgery for resectable disease.
However for lesions >3cm the local control rate decreases significantly. In a
study from Tanis et al local recurrence rate was 21.4% after RFA for lesions
>3cm, this is in accordance with the findings of Mulier et al where the local
recurrence rate after surgical RFA for liver tumours of 3-5cm was 21.7%. Local
recurrence after MWA for liver tumours >3cm showed a similar rate of 23%. Major
complications occur in 2-4% of patients receiving MWA.
By participating in the study, patients agree to undergo either SBRT or MWA.
For each participant, the method of treatment will be decided upon by
randomization. Pre-treatment screening will not be different from the standard
screening for these techniques and will not be an extra burden. Both SBRT and
MWA are considered safe and established treatment options for the target
population.
At the moment it is uncertain whether SBRT or thermal ablation is preferable
for intermediate size unresectable CRLM. No randomized controlled trials
comparing SBRT to MWA in CRLM have been conducted so far. If this study shows
that the efficacy of one technique is superior over the other this could lead
to a prolonged LTPFS for patients with CRLM, reducing the number of required
repeat procedures and potentially improving disease free and overall survival.