Factors Affecting LC Following SBRT for Liver Metastases

The utilization of stereotactic body radiation therapy for the treatment of liver metastasis has been widely studied and has demonstrated favorable local control outcomes. However, several predictive factors play a crucial role in the efficacy of stereotactic body radiation therapy, such as the number and size (volume) of metastatic liver lesions, the primary tumor site (histology), molecular biomarkers (e.g., KRAS and TP53 mutation), the use of systemic therapy prior to SBRT, the radiation dose, and the use of advanced technology and organ motion management during SBRT. These prognostic factors need to be considered when clinical trials are designed to evaluate the efficacy of SBRT for liver metastases.

liver
SBRT
oligometastases

1. Introduction

The liver is a common site of metastasis for several malignancies, most commonly from the primary tumor of the colorectum, breast, lung, stomach, esophagus, pancreas, and melanoma origin [1,2,3]. The most common source of liver metastasis is colorectal cancer (CRC), and up to 50% of these patients develop metastasis to the liver [3]. Additionally, an analysis of the Surveillance, Epidemiology, and End Results (SEER) database (2010–2015) showed that the overall incidence rate of liver metastases was 22.66 per 1,000,000 individuals [4].

In the past decades, local treatment of the liver oligometastasis has become increasingly common. The term “oligometastasis” was coined by Weichselbaum and Hellman in 1995 as a state between the absence of metastasis and the diffuse spread of disease [5]. In 2011, Weichselbaum and Hellman defined oligometastasis as metastasis limited in number and distribution as per standard diagnostic imaging scans [6,7]. In 2020, ESTRO-ASTRO published a consensus document defining oligometastatic disease as “1–5 metastatic lesions, a controlled primary tumor being optional, but where all metastatic sites must be safely treatable” [8].

Surgical resection, whenever feasible, is the mainstay treatment for medically operable patients with resectable liver oligometastases [9]. For those with liver metastases not suitable for surgical resection, localized treatments such as radiofrequency ablation (RFA), microwave ablation (MWA), or stereotactic body radiation therapy (SBRT) can be considered [1,10,11].

The Deutschen Gesellschaft für Radioonkologie (DEGRO) defined SBRT as a dedicated form of external beam radiation therapy with specific physical and biological characteristics. It is marked by a sharp transition in radiation dose outside the treated tumor while increasing the dose inside the specified tumor volume. This treatment modality is combined with image guidance techniques and advanced motion management to yield effective local treatment for liver oligometastases with a relatively low toxicity rate [11,12,13].

2. Tumor-Related Factors

2.1. Primary Tumor Type

Primary tumor origin and histology have been shown to impact outcomes amongst patients who received SBRT for liver metastases [12,52,53] (Table 2). The German group analyzed 474 patients with liver oligometastases from various histologies. Primary tumor origin was a significant predictor of LC. Metastases from CRC had a significantly worse LC rate at one year (67%) compared to breast cancer (91%), non-small cell lung cancer (88%), or other histologies (80%) [12]. Similarly, Ahmed et al. analyzed 372 liver metastases from various primary cancers following SBRT and reported significantly poorer LC rates for liver metastases of CRC origin, with one- and two-year LC rates of 79% and 59% for CRC lesions, compared to 100% for non-CRC lesions (p = 0.02) [54,55].

Table 2. Summarizes prospective and retrospective series on liver SBRT from various primary sites.

Primary Site	Author	Design	No. of Patient	No. of Liver Lesions	Dose	Median Follow Up (mo)	Median OS (mo)	1 y LC/OS	2 y LC/OS	1/2 y PFS
CRC	Hoyer et al. [33]	Phase 2	44	141	45 Gy/3 fx	52	19.2	67% OS	86% LC 38% OS	19% 2 y
	Vanderpool et al. [31]	Prospective	20	31	12.5–15 Gy/3 fx	26	34	100% OS 100% LC	74% LC 83% OS	-
	Chang et al. [27]	Retrospective	65	102	22–60 Gy/1–6 fx	14	-	72% OS 62% LC	45% LC 38% OS	-
	Py et al. [21]	Retrospective	67	99	37.5–54 Gy/3–5 fx	47	53	95.5% OS 86.6% LC	72.4% LC 81.4% OS	81% 1 y 54% 2 y
	Voglhuber et al. [30]	Retrospective	115	150	35 Gy/5 fr	11.4	20.4	72% OS 82% LC	82% LC 45% OS	20% 1 y 10% 2 y
	Yu et al. [26]	Retrospective	44	62	36–60 Gy/3–5 fr	31.8	-	96% OS	-	-
Esophageal	Li et al. [56]	Retrospective	8 (liver)	-	Median BED10 60 (39–90 Gy)	35	14	-	-	-
BTC	Franzese et al. [57]	Retrospective	21 (liver)	-	Median 45 Gy (24–75)/3–10 fx	14	13.7	58% OS 76.7% LC	71% LC 41% OS	36% 1 y 20%: 2 y
Pancreatic ADC	Lee et al. [58]	Retrospective	76	-	Median 50 Gy (40–50)/5 fx	10.9	8.5	38% OS 66% LC	-	7%: 1 y
GI NET	Hudson et al. [59]	Retrospective	25	53	Median 50 Gy/5 fr (25–60 Gy/3 fr)	14	-	92% LC	-	44%: 1 y
GYN	Laliscia et al. [60]	Retrospective	8 (liver)	-	24 Gy/1 fr, 27 Gy/3 fr	-	-	-	-	-
Breast	Milano et al. [61]	Prospective	14 (liver)	33 (curative)	-	-	-	-	76% OS	44%: 2 y
	Onal et al. [62]	Retrospective	22	29	54 Gy/3 fx	16	-	85% OS 100% LC	88% LC 57% OS	38%: 1 y 8%: 2 y
	Franzese et al. [63]	Retrospective	54	-	Median dose 60 Gy (30–75 Gy)/3 fx (3–6 fx)	26.2	-	95.5% OS 57.6% LC	41.6% LC 76.9% OS	38.7%:1 y 22%: 2 y
	Tan [64]	Retrospective	120 (24 liver)	29	30–60 Gy/3–5 fx	15.25	53.16	83.5% OS 89% LC	86.6% LC 70% OS	45%: 1 y 32%: 2 y
RCC & Melanoma	Stinauer et al. [65]	Retrospective	Melanoma 17 RCC 13	11	40–50 Gy/5 fr, 42–60 Gy/3 fx	28	24.3	88% LC	-	-
RCC & Melanoma	Grossman et al. [66]	Retrospective	16 RCC 15 melanoma	14 (liver)	Median SBRT 50 Gy, median fractional dose 5 Gy	-	10.8	94.7% LC	-	-
Melanoma	Franceschini et al. [67]	Retrospective	31 (8 liver)	11	50.25–75/3–6 fx	13	10.6	41%OS 96.6% LC	82.8% LC 21% OS	18.5%: 1 y 13.9%: 2 y

Abbreviations: CRC: colorectal cancer, BTC: biliary tract cancer, ADC: adenocarcinoma, GI NET: gastrointestinal neuroendocrine tumor, GYN: gynecological malignancies, RCC: renal cell carcinoma.

The impact of primary tumor type on LC of liver metastases after SBRT extends beyond the origin of the tumor and histologic subtype to the molecular phenotype of the cancer. Hong et al. analyzed the association of genetic alterations with LC in 89 patients (CRC being the most common primary) treated with proton-based liver-directed SBRT. They reported lower LC for lesions with KRAS mutation (one-year LC of 43% compared to 72%, p = 0.02) and reduced LC rates in cases with both KRAS and TP53 mutations (one-year LC of 20% compared to 69%, p = 0.001) [37,68].

2.2. Number, Size, and Volume of Metastatic Liver Lesions

Joo et al. reported that the number of treated liver lesions (one, two, or three) was a significant predictive factor for intrahepatic tumor control. Their findings suggest that an increasing number of treated sites was associated with a higher risk of reduced intrahepatic control [69]. Several trials had set a maximum tumor diameter of <6 cm for high-dose liver SBRT [2,15,25,70,71]. Doi et al. retrospectively reviewed the records of 24 patients with 39 metastatic liver tumors from CRC who were treated with SBRT. On multivariable analysis, a maximum tumor diameter ≤3 cm was significantly associated with better LC (p = 0.03) [72]. Similarly, Rusthoven et al. reported that for lesions with a maximum diameter ≤ 3 cm, 2-year LC was 100% compared to 77% for lesions > 3 cm (p = 0.015) [2]. Andratschke et al. reported that smaller treated metastatic tumor volume, as observed in the study where the GTV volume ranged from a minimum of 0.6 cc to a maximum of 699 cc (median volume of 27 cc), and the PTV volume ranged from a minimum of 4.5 cc to a maximum of 1074.0 cc (median volume of 71.3 cc), were found to be a significant predictor for LC (p < 0.001) [12]. Furthermore, Flamarique et al. reported that tumor volumes > 30 cc correlated with worsened two-year LC rates (90% vs. 34.5%) (p = 0.005) [73].

3. Treatment-Related Factors

3.1. Prior Liver-Directed Local Therapies

There is evidence to suggest that SBRT is a safe and well-tolerated treatment option with excellent LC rates for liver metastases that have been previously treated with other liver-directed therapy, including surgery, ablation, and transarterial chemoembolization (TACE) [74,75]. Moon et al. conducted a prospective single-arm trial to evaluate the safety and efficacy of liver SBRT in patients with or without prior liver-directed therapy. The study included a total of 30 patients, among whom 63% had liver metastases (47% had received prior liver-directed therapies, which included liver resection, TACE, and RFA). Out of the 30 patients, 28 underwent SBRT to a new lesion, and 2 received SBRT due to either a local recurrence or a sub-optimal response following TACE. The study did not find a statistically significant difference in LC between those who had previously undergone liver-directed therapies and those who had not (73% and 86% at one year, respectively, p = 0.70) [75].

3.2. Pre-SBRT Systemic Therapy

The surviving tumor cells, after systemic therapy, may develop a better ability to repair DNA damage, which results in a more radioresistant phenotype [52,76]. This may explain why metastatic cancer that was previously treated with adjuvant systemic therapy tends to be more aggressive [52,77]. Klement et al. analyzed 623 liver metastases in 464 patients who underwent SBRT treatment from any histology-proven primary solid tumor [52]. They found that patients who had received chemotherapy before SBRT had significantly reduced LC at two years compared to those who did not (58% vs. 83%, p = 0.04) [52]. Sheikh et al. conducted a multi-institutional retrospective analysis, which included 235 patients with a total of 381 CRC oligometastatic lesions treated with SBRT. On multivariable analysis, they found that receiving any systemic therapy before SBRT was linked to an increased risk of progression (p < 0.001) [55]. Furthermore, Andratschke et al. reported that patients who received systemic therapy before SBRT had worse LC rates compared to those who did not [12]. It is important to consider multiple factors when interpreting this correlation, as patients who received systemic therapy first might have had a higher burden of the disease. Future research employing novel biomarkers of disease burden (e.g., ctDNA analysis) is warranted.

3.3. SBRT Dose

Multiple prospective phase I/II trials for liver metastases have shown two-year LC rates ranging from 60% to 100% with different radiation dose and fractionation schedules. A phase II study by Scorsetti et al. treated 61 patients with liver metastases from different primary histologies with a dose of 75 Gy in three fractions showed a three-year LC of 78%, with no significant difference in LC based on histology (CRC vs. other) and size of lesion (>3 cm vs. <3 cm) when ultra-high dose SBRT was used [71]. For lesions measuring <3 cm, an ablative dose of 60 Gy delivered over three fractions yielded LC rates of 95% and 92% at one and two years, respectively [2]. For tumors measuring <6 cm, a higher ablative dose of 75 Gy given over three fractions achieved an LC rate of 94% [15]. A phase I/II dose escalation SBRT study by Rusthoven et al. showed two-year LC rates of 92% [2]. McPartlin et al. showed lower LC rates of 50% and 26% at one year and four years, respectively, in CRC liver metastases, likely due to lower SBRT dose used (median, minimum SBRT dose was 37.6 Gy (range, 22.7–62.1 Gy) in 6 fractions) [11].

Retrospective and modeling studies have shown improved LC with a high BED10 dose for liver metastases [78]. A single-institution retrospective study by Kok et al. showed a two-year LC of 90% with BED10 > 100 Gy10 vs. 60% with BED10 < 100 Gy10 [17]. Similar results were shown by Mahadevan et al. with BED10 > 100 Gy10 (wot-year LC 77.2% vs. 59.6%) in 427 patients with liver metastases [79]. Ohri et al. described better LC outcomes with BED10 > 100 Gy10, and the tumor control probability (TCP) modeling showed two-year LC increased to 76% at BED10 of 100 Gy10 and 90% with BED10 of 180 Gy10 [80]. While higher radiation doses could lead to better LC, it should be recognized that there is a tendency for high radiation doses to be prescribed for small tumors.

While the literature on single fraction SBRT is limited and early studies included both hepatocellular carcinoma (HCC) and liver metastases, more recent studies of single fraction SBRT for liver metastases with dose escalation to 35–40 Gy demonstrated two-year LC of 100% and four-year LC of 96.6% with no reported grade 3 toxicity [38,81]. Folkert et al. used a 35–40 Gy single fraction and applied the following constraints: max point dose 14 Gy, 12.4 Gy, 15.4 Gy, and 18.4 Gy to the spinal cord, stomach, and duodenum, jejunum, and colon, respectively. Furthermore, 700 mL of uninvolved liver received <9.1 Gy [38]. The results from these studies suggest that higher BED10 delivered in a single fraction can provide excellent LC with acceptable toxicities. Prospective randomized phase III trials are needed to further evaluate the efficacy and toxicity of single fraction ultra-high dose SBRT.

Despite the accumulating evidence that showed the association between SBRT dose and LC of liver metastases, it is not always possible in clinical practice to treat all liver metastases with very high dose SBRT mainly due to the proximity of the tumor to the central biliary tree or luminal structures (e.g., small/large bowel, stomach and duodenum). When selecting high-dose SBRT, it is also important to consider the volume of the uninvolved liver (i.e., whole liver minus gross tumor volume [GTV]) and the pre-SBRT liver function to avoid potential liver toxicity [2,11,12,73,82].

3.4. Advanced Organ Motion Management

Organ motion management is critical in liver SBRT in order to safely deliver ablative doses to the target while limiting the volume of normal tissue irradiated [83,84]. During liver SBRT, it has been reported that the internal motion of tumors can reach up to 39.5 mm (mean 17.6 mm) [85]. Various motion management mechanisms can be used, such as fiducial markers, abdominal compression, breath hold techniques, gating, and tumor tracking [83,86,87]. Imaging studies for motion measurement and evaluation include fluoroscopy, 4D CT, 4D cone-beam computer tomography scans, 2D cine MR, and 4D MR imaging [1,88,89,90]. Several studies have demonstrated that the use of advanced motion management techniques in liver SBRT is associated with improved LC of liver metastases [12,91]. In 2014, a report was published highlighting the importance of adequate respiratory motion management in SBRT for oligometastatic CRC patients. Results showed that metastases in moving organs (e.g., liver) exhibited an LC of 53% at 1 year compared to 79% for lymph nodes (p = 0.01) [92]. Klement et al. also reported that simple motion management techniques (such as free breathing and abdominal compression) predicted significantly lower tumor control probability [52].

This entry is adapted from the peer-reviewed paper 10.3390/curroncol30100667

© Text is available under the terms and conditions of the Creative Commons Attribution (CC BY) license; additional terms may apply. By using this site, you agree to the Terms and Conditions and Privacy Policy.