Treatment Outcomes from ⁶⁸Ga-PSMA PET/CT–Informed Salvage Radiation Treatment in Men with Rising PSA After Radical Prostatectomy: Prognostic Value of a Negative PSMA PET

DISCUSSION

Salvage prostate fossa RT is the current standard of care in men with biochemical failure after radical prostatectomy. It remains the last chance for cure in these men, with around 56% of men achieving complete biochemical response at 5 y after RT (2,3,11). Studies have shown that low serum PSA values after RP are correlated to significantly better treatment outcomes, supporting early SRT at a serum PSA of less than 0.2 ng/mL or less than 0.5 ng/mL (3). Treatment response drops off dramatically once the PSA rises above 1.0 ng/mL (4). However, a significant number of men, even with low PSA levels, do not respond to this treatment, presumably because of disease beyond the prostatic bed at the time of salvage. In a previous study, we demonstrated that up to 29% of men in this patient cohort have disease outside the prostate fossa at the time of imaging with PSMA PET/CT (5). We know that there is a high management impact with PSMA PET/CT in this patient population (7–9,12,13). However, what has not yet been evaluated is whether a finding of disease outside the prostate fossa on PSMA PET affects treatment outcomes.

This study shows that, in the cohort of men for whom salvage fossa radiotherapy remains standard of care, PSMA PET effectively stratifies men into those with a high (82.5%) versus low response (53%) to salvage RT. High treatment responders were men with negative PSMA PET results or with disease confined to the fossa, whereas low responders had PSMA avidity in nodes or distantly. Furthermore, this difference in treatment response was evident despite the fact that the RT fields were predominately limited in the high-treatment-response cohort, and involved more extensive radiotherapy fields in the low-treatment-response group. A PSMA PET result proved to be more predictive of treatment response to SRT than established clinical predictors, such as PSA level, Gleason score, pT stage, and surgical margin status.

PSMA PET/CT is a sensitive technique for identifying sites of recurrence in the post-RP biochemical failure setting at low PSA levels (6). A significant proportion of men with a PSA of less than 1.0 ng/mL will have a positive PSMA PET scan, with the detection rate of the scan dependent on PSA at the time of imaging (5,7,14–17). Men imaged with a PSA of between 0.05 and 0.5 ng/mL can expect to have a positive scan approximately 60% of the time, rising to 80% in the PSA range of 0.5–1.0 ng/mL. By contrast, we also know that PSMA PET underestimates the extent of disease. PSMA PET does not detect small-volume nodal deposits, with sensitivity of the technique dropping off sharply at sizes below 4 mm because of the inherent physical limitations of PET imaging (16,18). The sensitivity is further reduced in the region of the prostate fossa because of adjacent excreted activity in the bladder (19,20). The results of this study confirm that the extent of local recurrence may be underestimated by PSMA PET, and that this must be taken into consideration in interpretation of scan results.

We have demonstrated that the group with the highest treatment response to salvage fossa radiotherapy is those men with a negative PSMA PET scan. This may reflect the ability of PSMA PET to differentiate between disease phenotypes. Negative scans may identify patients with less aggressive disease who will benefit from local salvage and targeted treatment. Of the more than 56% of men with a negative PSMA PET who did not proceed to SRT, with the clinician electing to watch the PSA, up to 67% progressed biochemically. There were no significant differences in risk factors found between the men treated or not treated (Gleason score, T stage, surgical margins, PSA) with the decision to treat likely influenced by the negative finding on the PET scan. This high treatment response to SRT in men with a negative scan confirms that PSMA PET is not sufficiently sensitive to exclude microscopic disease recurrence around the prostate fossa. More significantly, it demonstrates that a high proportion of men with a negative scan respond to local (potentially curative) treatment, with low-volume recurrent disease amenable to RT. This has important implications for the clinical management of men with prostate cancer and biochemical failure.

The study also confirms that a negative or fossa-confined pretreatment PSMA PET was the strongest independent determinant of treatment response after SRT, even when PSA level at the time of scan, pT stage, Gleason score, or surgical margins were included in the analysis, reinforcing the paradigm that clinicians should recommend SRT even in men with a negative PSMA PET/CT. Although median PSA was lower in the men with negative PSMA PET, there was no difference in PSA at imaging between men with disease confined to the fossa and those with positive nodes or distant disease. The question as to what PSA level is optimum for PSMA PET imaging of men with biochemical failure has not been addressed by this study and requires well-designed prospective trials.

PSMA PET is being increasingly used in men with rising PSA after RP. A positive scan may significantly alter RT intent, doses, and volume. This study describes a significantly reduced treatment response in men with nodal involvement on PET or distant disease compared with either a negative scan or disease confined to the fossa, despite the more complex RT plans in those with disease beyond the fossa on PSMA PET. This highlights 2 points. First, it demonstrates the ability of PSMA PET to stratify patients into those who will respond well to standard fossa RT, compared with those who may require more extensive treatment. Second it raises questions regarding optimum treatment for those men with PSMA PET–positive findings outside the prostate fossa, targeted or systemic. This study has insufficient patient numbers to analyze the treatment response based on variations in RT. Further, the short follow-up interval would make this type of analysis futile.

This study had an overall 75% early response rate to SRT in men with rising PSA (0.05–1.0 ng/mL) after RP. This is lower than expected from the literature for biochemical response at this median PSA level (3). There are several possible reasons for this. First, all those who were treated with concurrent ADT were excluded from analysis. Further, the PSA decline after SRT continues for more than 12 mo, and median time in this study between SRT and follow-up PSA was just 10.5 mo. Our definition of treatment response involves low PSA levels not previously measurable before the widespread use of supersensitive PSA assays. The criteria we used for defining early treatment response (PSA < 0.1 ng/mL and >50% decline in PSA) to SRT were based on recent data demonstrating that a PSA nadir after SRT less than 0.1 ng/mL has prognostic implications (10) and attempted to define men who had a significant decline in PSA (>50% reduction) due to accurate targeting of disease. Finally, more than half the men with a negative PSMA scan, which we have shown in this study as predictive of treatment response, did not in fact receive RT. This may well have reduced our overall treatment response, and raises questions regarding optimal current practice for men undergoing PSMA-informed management.

A major limitation of this study is the short follow-up after treatment. With a median of 10.5 mo in the treated cohort of patients, it is difficult to compare treatment response with larger studies of SRT (measuring biochemical failure) in this population. Although it is not possible to extrapolate long-term findings from early treatment responses, it is most likely that the nonresponders (those with PSA rise despite SRT) will not be cured, and further follow-up is not required. Although the ability of PSMA PET to stratify treatment responses and characterize men with early biochemical failure has been demonstrated in this interim study, further follow-up will be undertaken to determine whether the stratification in biochemical failure persists up to 5 y after therapy, and whether PSMA targeted treatment is appropriate.

A further limitation of the study is the relatively low numbers of patients enrolled. This limits the study’s ability to evaluate appropriate PSA levels for imaging. Larger prospective studies with longer follow-up are needed to confirm that the prognostic value of a PSMA PET adds additional value to the predictive clinical findings included in the Stephenson nomogram and other predictors of salvage RT failure.