2 Clarke Drive
Suite 100
Cranbury, NJ 08512
© 2024 MJH Life Sciences™ and OncLive - Clinical Oncology News, Cancer Expert Insights. All rights reserved.
Stereotactic ablative body radiotherapy elicited high local control rates in patients with renal cell carcinoma who did not undergo surgical resection.
Stereotactic ablative body radiotherapy (SABR) elicited high rates of local control and was associated with an acceptable adverse effect (AE) profile and post-radiotherapy renal function in patients with primary renal cell carcinoma (RCC) who did not undergo surgical resection, meeting the primary end point of the phase 2 FASTRACK II trial (NCT02613819), results from which were published in Lancet Oncology.1
At a data cutoff of August 16, 2023, and a median follow-up of 43 months (interquartile range [IQR], 38-60), the 12-month local control rate was 100% (P < .0001). Investigators observed no local progression during the trial.
Although the standard of care for patients with primary RCC is surgery, older patients often have medical comorbidities that make them ineligible for surgical extirpation, particularly nephron-sparing surgery. Percutaneous thermal ablation is considered an alternative intervention for T1a renal masses. However, in tumors larger than 3 to 3.5 cm, the efficacy of this approach is reduced and the rate of complications increases.
“Patients with RCC who are not suitable for percutaneous thermal ablation or have T1b or larger tumors [> 4 cm] that are not medically operable have limited curative treatment options,” lead study author Shankar Siva, PhD, MBBS, FRANZCR, and coauthors, wrote in the paper. “Therefore, this non-surgical population is in need of an effective treatment alternative.”
Siva is the leader of the SABR program at the Peter MacCallum Cancer Centre in Victoria, Australia.
The single-center, phase 1 FASTRACK trial (NCT01676428) investigated SABR, a noninvasive curative treatment option that is feasible for T1a and T1b or greater tumors, in 37 patients with RCC who were unsuitable for surgery. In FASTRACK, at a median follow-up of 24 months, the respective 2-year rates of freedom from local progression, freedom from distant progression, and overall survival (OS) rates were 100%, 89%, and 92%.2
The international, nonrandomized FASTRACK II trial was conducted across 7 centers in Australia and 1 center in the Netherlands. Investigators enrolled patients 18 years of age or older with a biopsy-confirmed diagnosis of solitary primary RCC with only a single lesion who were medically inoperable; at high risk of surgical complications; or declined surgery.1 Patients needed to have an ECOG performance status (PS) of 0 to 2 and disease that warranted active treatment per multidisciplinary decision.
Patients were excluded if they had a pretreatment estimated glomerular filtration rate (eGFR) of less than 30 mL/min per 1.73 m2; received prior systemic therapies for RCC; underwent previous high-dose radiotherapy to an overlapping region; had tumors larger than 10 cm; and had direct contact between RCC and the bowel. Patients with untreated previous malignancies or a history of a previous malignancy within 2 years of screening; a horseshoe kidney; or visceral or bony metastatic disease were also excluded.
Patients received either a single fraction of SABR at 26 Gy for tumors up to 4 cm in maximum diameter, or 42 Gy in 3 fractions for tumors between 4 cm and 10 cm in maximum diameter.
This trial’s primary end point was local control per RECIST v1.1 criteria, measured from the date of SABR initiation to first evidence of local progression. Secondary end points included OS, cancer-specific survival, freedom from distant progression, safety, and renal function changes over time. A prespecified exploratory analysis aimed to investigate the effects of clinical and demographic factors on the 12-month eGFR.
Between July 28, 2016, and February 27, 2020, 70 patients were enrolled and treated. One patient enrolled and withdrew consent prior to treatment. Treated patients had a median age of 77 years (IQR, 70-82), a median body mass index of 32 kg/m2 (IQR, 27-38), a median Charlson comorbidity index of 7 (IQR, 5-8), and a median RENAL score of 8 (IQR, 4-11). Prior to enrollment, 70% of patients had documented serial growth on initial surveillance imaging. Most patients (70%) were male, and the median tumor size was 4.6 cm (IQR, 3.7-5.5). Thirty-four percent, 56%, 9%, and 1% of patients had T1a, T1b, T2a, and T3a disease, respectively. One patient had nodal involvement. ECOG PS was 0, 1, and 2 in 37%, 44%, and 19% of patients, respectively. Disease histology included clear cell (70%), papillary (17%), chromophobe (4%), oncocytic carcinoma (1%), and RCC not otherwise specified (7%). At baseline, 80%, 44%, and 39% of patients had hypertension, diabetes, and ischemic heart disease, respectively. Among patients evaluable for smoking status (n = 69), 11% were current smokers, and 49% were past smokers.
In total, 23 patients received single-fraction SABR in 26 Gy, and 47 patients were administered SABR at 42 Gy in 3 fractions. Treatment was administered as planned in 96% of patients. One patient required adaptive replanning after the first fraction because of a change in bowel position, and 2 patients required treatment interruption because of tumor bleeding and intratumoral hemorrhage.
The cancer-specific survival rate was 100%, and 97% of patients (95% CI, 89%-99%) had freedom from distant progression at 12 and 36 months from the start of SABR. The 12- and 36-month OS rates were 99% (95% CI, 90%-100%) and 82% (95% CI, 70%-89%), respectively.
The baseline mean eGFR was 61.1 mL/min per 1.73 m2 (95% CI, 56.5-65.6), which changed by –10.8 mL/min per 1.73 m2 (95% CI, –13.1 to –8.5) at 12 months and by –14.6 mL/min per 1.73 m2 (95% CI, –17.1 to –12.1) at 24 months from SABR initiation; changes plateaued thereafter. The 24-month eGFR estimate was 46.5 mL/min per 1.73 m2 (95% CI, 41.8-51.1).
The baseline split function estimate for the ipsilateral kidney was 50% (95% CI, 47%-53%). The 12- and 24-month ipsilateral kidney split function estimates were 36% (95% CI, 33%-39%) and 33% (95% CI, 30%-37%), respectively, and investigators observed a plateau onward from 24 months.
“The decline in renal function in FASTRACK II plateaued after 2 years, indicating a reversion to the background rate of decline due to chronic kidney disease factors after an initial drop secondary to SABR—a pattern that has been previously observed,” the study authors noted.
In a prespecified multivariable analysis of renal function predictors at 12 months, only baseline eGFR was associated with subsequent decline, with an average reduction of 8.4 mL/min per 1.73 m2 at 12 months per 10 mL/min per 1.73 m2 lower baseline eGFR (P < .0001). One patient with a 5.9-cm central tumor, a baseline eGFR of 34 mL/min per 1.73 m2, and a 12-month eGFR of 19 mL/min per 1.73 m2 required dialysis at 18 months with an eGFR of 7 mL/min per 1.73 m2.
Ten percent of patients experienced 1 or more grade 3 treatment-related AEs (TRAEs), including nausea and vomiting (4%); abdominal, flank, or tumor pain (6%); colonic obstruction (3%); and diarrhea (1%). Grade 1/2 TRAEs occurred in 74% of patients, and 16% of patients reported no TRAEs. No grade 4 AEs were reported, and no treatment- or cancer-related deaths occurred.
The study authors noted that key limitations of this trial included its small sample size and less mature follow-up compared with prospective surgical trials. Furthermore, FASTRACK II lacked a control group, preventing the assessment of SABR’s superiority, inferiority, or similarity to other treatment options.
“Given the absence of other potentially curative options for inoperable patients with larger tumors or a location not amenable to thermal ablation, SABR can be considered a proven modality,” the study authors emphasized.
The authors concluded by proposing that the FASTRACK II findings should be considered for escalation to a randomized, controlled trial of SABR vs surgery as the primary treatment modality in operable patients, and that future decision-making regarding the optional individualized treatment approach should be conducted in collaboration with the patients.
Related Content: