Two years after publication of level 1 evidence supporting focal RT boost, almost half of radiation oncologist respondents in our study have adopted this approach for their patients with prostate cancer. Subspecialists whose clinical practice focuses completely or nearly completely on genitourinary cancers were more likely to report use of focal boost. Nonetheless, a large proportion (39%) of these experts is not routinely using focal boost. Our results show a healthcare disparity exists where only patients seeing certain physicians will even be considered for focal boost.

It is important to note that rapid adoption of a new radiotherapy approach is not common, and widespread implementation of proven benefits is expected to take time. For example, despite multiple randomized clinical trials demonstrating noninferiority of moderate hypofractionation compared to conventional fractionation in prostate cancer treatment, adoption of hypofractionated regimens has been slow and variable across healthcare settings and individual physicians [4,5,6]. Such has been the case for other disease sites such as breast cancer, where moderate hypofractionation in adjuvant radiotherapy has been gradually integrated over nearly two decades [7, 8]. It is thus notable that a substantial proportion of the radiation oncologists surveyed are already routinely offering focal boost to their patients. Furthermore, there are certain barriers that have likely contributed to the slow uptake of hypofractionation, including a major financial disincentive induced by a fee-for-service payment model and concerns regarding the potential for increased toxicity [9, 10]. In contrast, the financial impact of the focal boost approach is less clear, and it has been shown to improve cancer outcomes without increasing toxicity.

Participants provided critical insight into barriers to their increased use of focal boost. Efforts to improve patient outcomes might address the most frequently cited barriers to adoption, including lack of access to high-quality MRI and concerns about accuracy of registration between MRI and CT images. Lack of access to high-quality MRI was more common in low-to-middle-income countries but remained a commonly cited barrier in high-income countries as well. Accurate registration between prostate MRI and CT is challenging due to variations in prostate appearance and pelvic anatomy, which can be attributed in part to movement of the prostate between scans [11]. While registration of the overall pelvis and pelvic bones is relatively straightforward, for focal boost, the goal is to register the prostate, which is difficult to precisely identify on CT and whose position is affected by variable bladder and rectal filling. Several methods have been developed to improve registration, including the use of implanted fiducials within the prostate, which were used in the FLAME trial [12]. Numerous automatic registration tools have also been developed that utilize machine learning and may aid in accurate and efficient registration of the prostate between MRI and planning CT [13,14,15].

Another commonly perceived barrier, especially among generalists, is the lack of training to identify prostate tumors on MRI. Lesion identification is not a common component of radiation oncology training. In the FLAME trial, oncologists had the assistance of expert radiologists for each case. Improved technology can also help to overcome this barrier. For example, a novel prostate cancer MRI biomarker (called the Restriction Spectrum Imaging restriction score, or RSIrs) makes it easier to see clinically significant cancer [16,17,18,19]. RSIrs can be obtained on clinical scanners with a 2–4-min diffusion-weighted acquisition, in addition to anatomic T2-weighted MRI. In a prospective study, use of RSIrs markedly improved radiation oncologists’ accuracy in identifying prostate tumors [20].

Some participants expressed doubt about the benefit of focal boost, despite the results of the FLAME trial. The initial FLAME paper reported only a disease-free survival advantage of focal boost, with an increase from 85 to 92% at 5 years compared to the standard arm [1]. Some physicians may be unaware of the regional/distant metastasis-free survival advantage described in a subsequent publication, where the regional and distant metastatic failure rate was reduced by nearly half in the focal boost arm [2]. Additional ongoing trials may also corroborate the FLAME results, encouraging adoption.

Other participants expressed valid concerns about the potential for increased toxicity. In the FLAME trial, the cumulative incidence of late grade ≥ 2 GU toxicity was 23% in the standard arm and 28% in the focal boost arm, whereas that of late grade late grade ≥ 2 GI toxicity was 12% in the standard arm and 13% in the focal boost arm [1]. These differences were small and not statistically significant. The focal boost dose on that trial was escalated up to 95 Gy but only to the extent feasible while meeting dose constraints to normal tissues, suggesting that prioritizing organ at risk constraints allows focal boost doses to be delivered safely yet effectively.

Additionally, some participants had concerns about focal boost in the setting of larger doses per fraction than used in FLAME. Data on this topic are emerging [21]. Hypo-FLAME was a phase II, single-arm study of ultra-hypofractionation (5 weekly fractions) with a focal boost up to 50 Gy. This study incorporated a urethral dose constraint, as recommended by Groen et al., and found acceptable toxicity [22]. Phase III trial evidence is not available for focal boost with ultra-hypofractionated regimens. DELINEATE was a single-center phase II trial in the UK that recently demonstrated safety and feasibility of using a focal boost of 67 Gy in 20 fractions or 82 Gy in 37 fractions. The efficacy and toxicity rates at 5 years were comparable to those in published trials, including FLAME. PIVOTALboost is an ongoing phase III randomized trial in the UK evaluating a focal boost of 67 Gy in a 20-fraction hypofractionated regimen [23]. Ideal constraints are still under investigation, and some patients may not be good candidates for boosting [24]. On the other hand, if hypofractionation is considered a key barrier to boosting, the logistic advantages of hypofractionation must be weighed against the oncologic benefit of focal boost.

Although not directly addressed in our questionnaire, implementation strategies are also important in increasing adoption of new techniques. Implementation science is the study of methods that seek to promote the uptake of evidence-based practices. System-wide changes rely on structural and organizational support to enable the initiation and expansion of implementation strategies. This includes clinical networks, which are comprised of healthcare leaders who aim to identify areas of improvement in clinical care and service delivery and advocate for system-wide change using a collaborative approach [25]. Clinician-centered educational programs targeting behavioral change have the potential to improve guideline adherence at a regional level. For example, one such educational program reduced unnecessary imaging in men with localized, low-risk prostate cancer [26]. At the provider level, the clinical champion approach is especially effective at impacting individual physician behavior. Strategies like these aim to accelerate the integration of research findings into clinical practice by identifying and addressing contextual barriers that contribute to the evidence-practice gap [27].

The FLAME trial applied standard clinical techniques in widespread use today. However, additional technologies may play a role in expanding the feasibility of focal RT boost. For example, a posterior tumor may not be amenable to a robust focal tumor boost without violating rectal dose constraints, but placement of a hydrogel spacer could yield more favorable dosimetry for the focal boost. Similarly, adaptive planning and MR-linac platforms could facilitate tighter planning margins and/or more accurate focal boosting. Focal-only brachytherapy boost and intensity modulated proton therapy boost are also possible areas for further study.

Limitations of this study include self-reported practice patterns and a sample of convenience, which led to the overrepresentation of physicians at academic medical centers and of genitourinary subspecialists. The questionnaire was also very brief to encourage participation and does not provide a comprehensive picture of all aspects of practice patterns, including how physicians who do offer focal boost select candidates for this approach or how they identify the target volumes.

In conclusion, in responses from over 250 international radiation oncologists, we found that almost half are routinely offering focal RT boost. Of note, there was overrepresentation in our study of subspecialists in genitourinary cancers, who might be earlier adopters. Based on commonly cited barriers, further adoption of focal RT boost might be aided by increased access to high-quality MRI, better registration algorithms of MRI to CT simulation images, more clinical data (especially for larger fraction sizes), physician education on benefit-to-harm ratio, and physician training on how to contour prostate lesions on MRI. Addressing these barriers would likely increase the adoption of focal RT boost and improve the efficacy of RT for more patients with prostate cancer.