Prognostic value of ctDNA fraction in mCRPC: results from the ProBio trial

International Investigator’s Meeting, October 7-8, 2024
ProBio sponsors

Alessio Crippa

Outline

 

  • Background and aims
     

  • Patients population and methods
     

  • Results
     

  • Conclusions
     

  • Q&A

Background and aims

Background

 

  • ctDNA fraction has shown promises as prognostic factors.

  • Most of the evidence comes from retrospective studies.
     

  • Methodological limitations:

    1. Dichotomization (detectable vs undetectable) or categories with arbitrary cut-points (i.e. <2.5%, 2.5%-30%, >30%).
    2. Unclear benefits in terms of survival (HR for different endpoints).

Aim

 

  • Aims of this study:

    • Elucidate the prognostic role of undectectable ctDNA in terms of time no-longer clinically benefitting (NLCB) and Overall Survival (OS).
    • Estimate the dose-response relationship between increasing ctDNA fraction levels and NLCB, and OS.

Patients population and methods

Patients population

Methods (1)

 

Prognostic role of undetectable ctDNA.  

  1. Bayesian Weibull accelerated failure time models.

  2. Survival Time Ratio (STR) as treatment effect.

  3. Adjustment by multiple prognostic factors: received therapy, ECOG, location of metastases, timing of metastatic disease, treatment line, analgesics use, previous ARPI and taxanes, log PSA, log lactate dehydrogenase, albumin, alkaline phosphatase and haemoglobin.

  4. Differences in treatment effects by received therapy.

Methods (2)

 

Dose-response relationship  

  1. As before, Bayesian survival models adjusted by confounders.
     

  2. Spike at zero model:

  • Undetectable ctDNA as separate group.
  • Splines for non-linear dose-response curves.

Prognostic role of undetectable ctDNA

Prognostic role of undetectable ctDNA

 

Group

n

Events

Median Time

Crude STR
(90% CrI)

Adjsuted STR
(90% CrI)

Fully Adjusted STR
(90% CrI)

No-longer clinically benefitting

Undetectable ctDNA

81

48

21.2 (17.9, 25.3)

Referent

Referent

Referent

Detectable ctDNA
(Physician's Choice)

56

50

8.5 (7.0, 10.3)

0.40 (0.31, 0.51)

0.52 (0.42, 0.66)

0.61 (0.48, 0.78)

Detectable ctDNA
(Investigational Arms)

83

71

9.7 (8.2, 11.4)

0.46 (0.36, 0.57)

0.61 (0.49, 0.76)

0.68 (0.54, 0.86)

Overall survival

Undetectable ctDNA

81

25

45.6 (38.6, 55.1)

Referent

Referent

Referent

Detectable ctDNA
(Physician's Choice)

56

39

22.2 (19.2, 26.1)

0.49 (0.38, 0.61)

0.53 (0.42, 0.67)

0.74 (0.57, 0.93)

Detectable ctDNA
(Investigational Arms)

83

51

25.8 (22.7, 29.7)

0.57 (0.45, 0.70)

0.62 (0.50, 0.77)

0.80 (0.64, 0.99)

Differential treatment effect

Dose-response (spike-at-zero)

For a metachronous patient receiving ARPI after ADT + docetaxel, with ECOG 0 and analgesics, metastases in bone and LN, and median PSA (14.8), LDH (3.4), albubin (41), ALP (1.6), HB (132).

Summary and conclusions

  • Patients with detectable ctDNA had a 60% and 40% shorter time to NLCB and OS.

  • We demonstrated a linear negative association between increasing ctDNA levels and NLCB, and OS.

  • Results were consistent in multivariate analyses.
     

  • mCRPC patients with undetectable ctDNA had a more favourable prognosis compared to those with detectable ctDNA.

  • Alternative therapeutic approaches can be more suitable for patients with undetectable ctDNA.

Prognostic value of ctDNA fraction