Advice

Clinical and technical evidence

Clinical and technical evidence

A literature search was carried out for this briefing in accordance with the interim process and methods statement. This briefing includes the most relevant or best available published evidence relating to the clinical effectiveness of the technology. Further information about how the evidence for this briefing was selected is available on request by contacting mibs@nice.org.uk.

Published evidence

The evidence base for Axumin includes over 20 studies and case reports identified as being potentially relevant. One meta-analysis and 5 studies (2 randomised trials and 3 non-randomised studies), involving over 1,200 patients are included in this briefing based on relevance and quality of evidence. Table 1 summarises the clinical evidence, as well as its strengths and limitations.

Overall assessment of the evidence

Overall, the evidence suggests that Axumin positron emission tomography (PET)/CT can detect prostate cancer recurrence with good diagnostic accuracy. The technology has been shown to detect local and distant areas of recurrence across a wide range of prostate-specific antigen (PSA) levels and is generally well tolerated.

Evidence shows that Axumin PET/CT can detect recurrence at low PSA levels (below 1 ng/ml) and that detection rates vary with PSA levels, improving as levels increase. In the LOCATE study, a detection rate of 31% was reported for patients with the lowest PSA (0 to 0.5 ng/ml), which rose to 50% and 66% for PSA ranges of 0.5 to 1 ng/ml and 1.0 to 2.0 ng/ml respectively. In Bach-Gansmo et al. 2017, the detection rate in the lowest PSA quartile (below 0.79 ng/ml) was 41%; 30% of patients had recurrence detected outside of the prostate. Both of these studies included a mixed post-therapy population (men with biochemical relapse after radical prostatectomy, or radiotherapy or both). In another study involving men with biochemical relapse after radical prostatectomy only, the Axumin PET/CT detection rate was 72% when the PSA level was less than 1 ng/mL, but increased to 83% and 100% when the PSA levels were 1 to 2 ng/ml or 2 ng/ml and over respectively (Akin-Akintayo et al. 2017).

The evidence reports that Axumin use frequently results in major changes to patients' management plans compared with standard imaging (abdominopelvic CT, MRI or bone scans), particularly in the extent of planned salvage radiotherapy and the volumes targeted for radiotherapy. However, it is not certain from the available evidence that changes in management lead to improved outcomes for patients. This is an uncertainty common to all imaging modalities used to locate areas of prostate cancer recurrence. Also, with the exception of the FALCON study, all available evidence on changes to clinical management plans come from studies involving US centres, which may not reflect NHS practice. Longer-term studies or patient registries would be needed to assess the effect of Axumin-guided therapy on clinical outcomes such as earlier detection of recurrence, subsequent treatment and progression-free survival. Understanding the performance of Axumin PET/CT in different populations, such as in people grouped by previous treatment, Gleason score and PSA doubling times, may provide information that could help inform the selection of patients for scanning.

In addition to the studies summarised in the table 1, 4 single-centre prospective studies (Nanni et al. 2013; Nanni et al. 2014; Nanni et al. 2015; Nanni et al. 2016), and a case-series involving 10 patients (Calais et al. 2018) have compared the diagnostic performance of Axumin to 11C‑choline and prostate-specific membrane antigen (68Ga‑PSMA) respectively. Overall, data from these studies suggest that Axumin may be superior to 11C‑choline but inferior to 68Ga‑PSMA in detecting prostate cancer recurrence. No head-to-head data comparing Axumin with 18F choline were identified.

Table 1 Summary of selected studies

Andriole et al. (2019)

Study size, design and location

Prospective, open-label, multicentre, interventional LOCATE study looking at whether 18F‑fluciclovine-PET/CT testing changes clinical management in 213 men who had had curative-intent treatment but who were suspected as having recurrence based on rising PSA levels. The study was carried out across 17 US centres.

Intervention and comparator(s)

Intervention: 18F‑fluciclovine-PET/CT.

No comparator.

Key outcomes

18F‑fluciclovine-PET/CT detected 1 or more sites of recurrence in 57% of men with BCR. Overall, 59% (126/213) of patients had a change in management; 78% (98/126) of these were 'major' changes. The most frequent major changes were from salvage or non-curative systematic therapy to watchful waiting (25%), from non-curative systematic therapy to salvage therapy (24%) and from salvage therapy to non-curative systemic therapy (9%). At 6 months, 63% of patients had treatment that was concurrent with post-scan plans, a clinically important difference was found for 37%.

Strengths and limitations

Study included a relatively large number of patients across 15 centres, reducing bias and increasing generalisability. The study included a 6‑month follow-up. It included US centres only, so results may not be generalisable to a UK NHS setting. Imaging results were not routinely confirmed with histological findings. The study was sponsored by the company.

Teoh et al. (2018)

Study size, design and location

Prospective, open-label, multicentre, interventional FALCON study involving 85 men being considered for curative-intent salvage therapy after initial radical therapy (pre-planned interim analysis of the first 85 patients). The study was carried out across 6 UK centres.

Intervention and comparator(s)

Intervention: 18F‑fluciclovine PET/CT.

No comparator.

Key outcomes

Most of the men in this pre-planned interim analysis had previously had radical prostatectomy (65.9%), with 27 men having had salvage radiotherapy (± other therapy). After 18F‑fluciclovine PET/CT imaging, 61.2% of those imaged had a change in management, suggesting 18F‑fluciclovine PET/CT has an effect on clinical decisions for men with a first BCR of prostate cancer after curative-intent primary therapy. Recruitment was subsequently stopped as the pre-specified cut-off for efficacy (>45 treatment changes) was met.

Strengths and limitations

This was a multicentre study involving 6 UK sites, so results are likely to be generalisable to NHS practice. It was a planned interim analysis of the first 85 men published as a conference abstract only, so full study details and patient demographics were not available. Study was sponsored by the company.

Bach-Gansmo et al. (2017)

Study size, design and location

Multicentre, retrospective, observational BED‑001 study involving 596 patients who had 18F‑fluciclovine PET/CT at 4 centres (1 in US, 1 in Italy and 2 in Norway).

Intervention and comparator(s)

Intervention: 18F‑fluciclovine PET/CT.

No comparator.

Key outcomes

At a subject level, 18F‑fluciclovine PET/CT showed a detection rate of 67.7% (403/595 scans). At a regional level, positive findings were detected in the prostate/bed in 38.7% and in pelvic lymph nodes in 32.6% of scans. Metastatic involvement outside the lymph nodes was detected in 26.2% of scans. Subject level detection rate in those in the lowest quartile for baseline PSA (0.79 ng/ml or less) was 41.4%. The positive predictive value of 18F‑fluciclovine PET/CT for all sampled lesions was 62.2%, and was 92.3% and 71.8% for extraprostatic and prostate/bed involvement respectively. Treatment emergent adverse events were experienced by 5.4% of patients but none were considered adverse reactions to 18F‑fluciclovine. The safety profile was not noticeably altered after repeat administration.

Strengths and limitations

Study included a large number of people from 4 clinical centres. It involved non-UK centres only so results may not be transferable to NHS clinical practice. The retrospective nature of the study meant that the use of comparative imaging and biopsy technique were not standardised, information on change in treatment was not be systematically captured, and findings could not be confirmed by histology in all patients. Only adverse events occurring within the first 35 days after administration were determined from site records.

Jani et al. (2017)

Study size, design and location

Prospective, single-centre randomised, controlled clinical trial involving 96 patients with rising levels of PSA after prostatectomy being considered for salvage radiotherapy.

Intervention and comparator(s)

Intervention: 18F‑fluciclovine PET/CT (n=49).

Comparator: Radiation therapy based on conventional imaging (bone scan and abdominopelvic CT and/or MRI scan; n=47).

Key outcomes

With the exception of PTV2, all post-registration volumes were statistically significantly larger than the corresponding pre-registration volumes, suggesting that, in most cases, including information from 18F‑fluciclovine PET/CT in the treatment planning process may lead to larger volumes targeted for salvage therapy. This was associated with higher doses of radiotherapy (40 Gy and 60 Gy) to the penile bulb (p=0.001 and p=0.002 respectively), but no statistically significant difference in rectal or bladder doses. Acute toxicity results were not statistically significantly different between the control and experimental arms and no acute grade 3, 4 or 5 toxicity was seen in the experimental arm, suggesting treatment to the modified clinical target is tolerable.

Strengths and limitations

Randomised trial with intention-to-treat design, helping to remove selection bias and avoid the effects of loss to follow-up. This was a planned analysis of secondary end points for the first 96 patients (accrual goal is 162), so the current follow-up only shows provider-reported acute toxicity outcomes; no information on disease-free survival or long-term toxicity can be determined. Single-centre study conducted in the US, results may not be generalisable to a UK NHS setting. Study was sponsored by the company.

Akin-Akintayo et al. (2017)

Study size, design and location

Randomised, prospective, intention-to-treat clinical trial involving men with PSA failure after radical prostatectomy. 87 men were recruited and 44 were randomised to the intervention.

Intervention and comparator(s)

Intervention:18F‑fluciclovine PET/CT.

Comparator: standard care.

Key outcomes

2 patients dropped out before PET/CT scanning.

81.0% (34/42) of patients had a positive results on 18F‑fluciclovine PET/CT. All 42 patients who had 18F-fluciclovine PET/CT were initially planned for radiotherapy. After 18F‑fluciclovine PET/CT findings, radiotherapy decisions were changed in 40.5% patients (17/42). 4.8% (2/42) of patients had radiotherapy decisions withdrawn because of evidence of extra-pelvic disease. Of the remaining patients, 37.5% (15/40) had radiotherapy fields changed; with 73.3% (11/15) of these fields increased from prostate bed only to both prostate and pelvis, and 26.7% (4/15) reduced from both prostate bed and pelvis to the prostate bed only. Changes in radiotherapy field and overall radiotherapy decision were both statistically significant (p<0.001) with 18F‑fluciclovine PET/CT, but change in the decision to offer radiotherapy or not was not statistically significant (p=0.15).

Strengths and limitations

This was a planned interim analysis of a secondary end point and the study had not reached final accrual goal (81 patients in each arm). Pre-fluciclovine radiotherapy decisions would have been influenced by a number of factors including patient history, pathology and PSA trajectory. However, several radiotherapy providers were used, meaning pre-fluciclovine decisions are likely to be representative of those made in clinical practice. Single-centre study conducted in the US, so results may not be generalisable to a UK NHS setting. The study centre was given cassettes of 18F‑fluciclovine by the company.

Ren et al. (2016)

Study size, design and location

Systematic review and meta-analysis of published data about the performance of 18F‑fluciclovine PET/CT in the diagnosis of recurrent prostate cancer. 6 studies were included, involving a total of 251 patients with suspected prostate cancer recurrence.

Intervention and comparator(s)

Intervention: 18F‑fluciclovine PET/CT.

Comparators: 11C‑choline PET/CT, 111In-capromab pendetide SPECT/CT, T2W MRI.

Key outcomes

18F‑fluciclovine PET/CT had an 87% pooled sensitivity, and a 66% pooled specificity. On a per-patient-based analysis in detecting prostate cancer recurrence, 18F‑fluciclovine PET/CT had an area under the receiver-operating characteristic curve of 0.93.

Strengths and limitations

Only 6 studies were included in this meta-analysis; none of which were randomised controlled trials and most included a relatively small number of patients. All included studies involved histology, follow-up methods or both, and the methodology of these studies were evaluated and rated by 3 independent reviewers. Included studies were statistically heterogeneous in their estimates of specificity, possibly because of differences in methodology between the studies (for example, prospective versus retrospective design).

Abbreviations: BCR, biochemical recurrence; PET, positron emission tomography; PSA, prostate-specific antigen; SPECT, single photon emission computed tomography; T2W, T2-weighted.

Recent and ongoing studies