Rapid PSA Decline and Resistance to Abiraterone Acetate
Rapid PSA Decline and Resistance to Abiraterone Acetate
To the best of our knowledge, this is the first study aimed at identifying factors predicting primary NHA resistance. In our experience of everyday clinical practice in an unselected CRPC population treated with AA and ENZ in two successive named patient programs, the only indicator of possible primary resistance was the lack of a major biochemical response (i.e., a reduction in PSA levels of more than 50% in comparison with baseline) within 1 month of starting treatment. Probably owing to the low number of patients involved in this analysis, the role of this parameters was clearly demonstrated only in the cumulative analysis of the sample: in the ENZ group, PSA reduction more than 50% at 1 month after treatment start resulted significant in univariate analysis, but not in multivariate analysis, while in the AA group, despite the similar magnitude of effect due to this parameter, its role was not confirmed (Figure 1). However, a clear difference in PFS and OS was observed by 1-month PSA ≥50% reduction using a 1-month landmark analysis.
After the publication of the results of the NHA pivotal trials in 2011 and 2012, some preliminary studies of predictive and/or prognostic factors were published concerning patients receiving AA, whereas no data are available concerning ENZ. Loriot et al. found that the previous duration of prostate cancer sensitivity to androgen deprivation therapy (classified as ≥16 months and <16 months) significantly and strongly predicted a PSA response (58 vs 18%; p = 0.01) and PFS in CRPC patients receiving subsequent endocrine manipulations; however, their analysis included not only patients treated with AA, but also those receiving older generation agents, such as ketoconazole-hydrocortisone, dexamethasone and bicalutamide, that are known to interfere with hormonal mechanisms. Two further preliminary studies specifically concerning AA showed that a Gleason score of >7 and >1 chemotherapy line before AA were independent predictive factors of a nonresponse to the drug, and that primary resistance to docetaxel predicts a lack of AA anti-tumoral activity.
The pivotal studies clearly showed that some patients treated with NHAs have primary resistance, which means that the therapy fails to reduce PSA levels or lead to an objective response when measured after the benchmark 12-week period.
Primary resistance to NHAs is a clinically relevant landscape not only because it leads to rapid progression and the need for further treatment modalities, but also because it plays a central role in relation to a global sequencing strategy that includes not only AA and ENZ, but also cabazitaxel and other already available or forthcoming therapeutic options (radium 223, cabozantinib and orteronel).
While awaiting factors capable of predicting the overall response to treatment, it could be clinically useful to identify factors that may help to identify patients at risk of primary resistance.
Interesting data on predictive value of circulating tumor cells assessment are available in patients treated with AA, but this methods could be not easily translated in the clinical practice.
In our experience, the absence of a major biochemical response during the first treatment month was the only indicator of possible primary NHA resistance. The patients who failed to achieve a ≥50% reduction in PSA levels within this time had a nearly 50% probability of progression within 3 months as against the 18% probability of those who did (in AA patients 53 and 25%, in ENZ patients 42 and 11%, respectively).
The role of PSA levels as a surrogate marker of CRPC clinical outcomes, such as PFS or OS, has been widely debated and remains controversial. The role of PSA-related parameters has been investigated in terms of their prognostic and/or predictive value during both the pretreatment and on-therapy period. In the case of patients treated with first-line chemotherapy, pretreatment PSA levels and doubling times have been considered independent risk factors for mortality over time. However, it could be interesting to test PSA kinetics during therapy in an attempt to update the prognosis of individual patients over time although, in this case, the main problem is whether changes in PSA levels over the time reflect real changes in disease burden.
In the case of chemotherapy, conflicting results have been obtained since a ≥30% reduction in PSA levels within 3 months of starting therapy demonstrated the highest degree of surrogacy for OS with docetaxel, but it did not with cabazitaxel. Nevertheless, the relationship between the speed of the decline in PSA levels and survival outcomes in patients treated with docetaxel has also been confirmed in everyday clinical practice. It has been demonstrated that a model based on a classification of on-treatment PSA kinetics (responders, late progressors and initial progressors) fits well with working group-defined progression criteria. Our experience started from a similar (albeit inverted) approach insofar as we found that early responders are at less risk of progressing within 3 treatment months and, consequently, at less risk of showing primary resistance to NHAs.
Discussion
To the best of our knowledge, this is the first study aimed at identifying factors predicting primary NHA resistance. In our experience of everyday clinical practice in an unselected CRPC population treated with AA and ENZ in two successive named patient programs, the only indicator of possible primary resistance was the lack of a major biochemical response (i.e., a reduction in PSA levels of more than 50% in comparison with baseline) within 1 month of starting treatment. Probably owing to the low number of patients involved in this analysis, the role of this parameters was clearly demonstrated only in the cumulative analysis of the sample: in the ENZ group, PSA reduction more than 50% at 1 month after treatment start resulted significant in univariate analysis, but not in multivariate analysis, while in the AA group, despite the similar magnitude of effect due to this parameter, its role was not confirmed (Figure 1). However, a clear difference in PFS and OS was observed by 1-month PSA ≥50% reduction using a 1-month landmark analysis.
After the publication of the results of the NHA pivotal trials in 2011 and 2012, some preliminary studies of predictive and/or prognostic factors were published concerning patients receiving AA, whereas no data are available concerning ENZ. Loriot et al. found that the previous duration of prostate cancer sensitivity to androgen deprivation therapy (classified as ≥16 months and <16 months) significantly and strongly predicted a PSA response (58 vs 18%; p = 0.01) and PFS in CRPC patients receiving subsequent endocrine manipulations; however, their analysis included not only patients treated with AA, but also those receiving older generation agents, such as ketoconazole-hydrocortisone, dexamethasone and bicalutamide, that are known to interfere with hormonal mechanisms. Two further preliminary studies specifically concerning AA showed that a Gleason score of >7 and >1 chemotherapy line before AA were independent predictive factors of a nonresponse to the drug, and that primary resistance to docetaxel predicts a lack of AA anti-tumoral activity.
The pivotal studies clearly showed that some patients treated with NHAs have primary resistance, which means that the therapy fails to reduce PSA levels or lead to an objective response when measured after the benchmark 12-week period.
Primary resistance to NHAs is a clinically relevant landscape not only because it leads to rapid progression and the need for further treatment modalities, but also because it plays a central role in relation to a global sequencing strategy that includes not only AA and ENZ, but also cabazitaxel and other already available or forthcoming therapeutic options (radium 223, cabozantinib and orteronel).
While awaiting factors capable of predicting the overall response to treatment, it could be clinically useful to identify factors that may help to identify patients at risk of primary resistance.
Interesting data on predictive value of circulating tumor cells assessment are available in patients treated with AA, but this methods could be not easily translated in the clinical practice.
In our experience, the absence of a major biochemical response during the first treatment month was the only indicator of possible primary NHA resistance. The patients who failed to achieve a ≥50% reduction in PSA levels within this time had a nearly 50% probability of progression within 3 months as against the 18% probability of those who did (in AA patients 53 and 25%, in ENZ patients 42 and 11%, respectively).
The role of PSA levels as a surrogate marker of CRPC clinical outcomes, such as PFS or OS, has been widely debated and remains controversial. The role of PSA-related parameters has been investigated in terms of their prognostic and/or predictive value during both the pretreatment and on-therapy period. In the case of patients treated with first-line chemotherapy, pretreatment PSA levels and doubling times have been considered independent risk factors for mortality over time. However, it could be interesting to test PSA kinetics during therapy in an attempt to update the prognosis of individual patients over time although, in this case, the main problem is whether changes in PSA levels over the time reflect real changes in disease burden.
In the case of chemotherapy, conflicting results have been obtained since a ≥30% reduction in PSA levels within 3 months of starting therapy demonstrated the highest degree of surrogacy for OS with docetaxel, but it did not with cabazitaxel. Nevertheless, the relationship between the speed of the decline in PSA levels and survival outcomes in patients treated with docetaxel has also been confirmed in everyday clinical practice. It has been demonstrated that a model based on a classification of on-treatment PSA kinetics (responders, late progressors and initial progressors) fits well with working group-defined progression criteria. Our experience started from a similar (albeit inverted) approach insofar as we found that early responders are at less risk of progressing within 3 treatment months and, consequently, at less risk of showing primary resistance to NHAs.
Source...