They recently completed an institutional study of patients with late relapse (LR) at MSKCC using their prospectively maintained database (1990-2017). They identified 157 patients with LR; 143 NSGCT and 14 seminomas. Of these patients, 30 were chemo-naïve. Patient demographics are below:
He then sequentially went through their main findings, which I summarize below.
1. Cancer-specific survival
- 5-year CSS NSGCT: 69.1% (but median time to LR was 99.6 month)
- 5-year CSS seminoma: 78.8% (but median time to LR was 59 months)
2. The pre-and-post LR management of NSGCT
- This was a busy slide, but important to note that all patients are susceptible to LR
- 19 patients just had orchiectomy and had LR on surveillance
- 48 patients required re-do operations after initial surgery for LR – so repeat surgeries are often necessary
- Patients who underwent surgery generally did better: 5-year CSS ~75% vs. 29% (no surgery)
- 95% were nodal metastases, and 88% were retroperitoneal – primary site of LR
- 28% had visceral metastases
4. Predictors of CSS
- While there were many preclinical univariate predictors of CSS, the only predictor of MV analysis was prior chemotherapy receipt – patients who had received prior CTX did much worse! (HR 5.6, p<0.05)
- Patients not treated with surgery post-LR and those with viable cancer and SMT fared much worse than surgically treated patients and patients with teratoma
5. Surgery for LR
- 123 (86%) underwent surgery for LR; 48 (58.5%) required a re-do RPLND
- Adjunctive procedures were often required
6. Histology at LR surgery
- Yolk sac (50%), teratoma (68.6%), somatic transformation (21%) were the most prominent histology at the time of LR; fibrosis 3.7%
7. Sites of relapse and treatment were associated with survival
8. Time to LR
- Occurred as late as 38 years later – so lifelong monitoring is required
- CSS was not dependent on time to LR – no time cutoff identified as a predictor of better/worse CSS
- However, SMT was noted to develop later than viable tumor or teratoma
What is the etiology of LR? In Sheinfeld’s opinion, it is the result of a vulnerable retroperitoneum – which is due to (depending on initial presentation):
He then reviewed some additional studies to highlight the following points:
1. Even Stage 1 tumors are at risk for LR (2.5-5% in the Mortensen EU 2016 paper)
2. Clinical T1a vs. T1b – T1a tend to have a higher rate of LR; but the LR in T1b tend to be more aggressive. (Kollmannsberger JCO 2014)
3. Template dissection is a disservice to the patient – MSKCC outcomes improved once they abandoned template dissections in 1998 (Stephenson JCO 2005, Carver JCO 2007)
4. “Relapsers relapse” – unfortunately, the patients that relapsed early are more likely to relapse late!
His take-home points are:
Presented by: Joel Sheinfeld, MD, Memorial Sloan Kettering Cancer Center, NYC, New York
Written By: Thenappan Chandrasekar, MD, Clinical Fellow, University of Toronto Twitter: @tchandra_uromd at the 73rd Canadian Urological Association Annual Meeting - June 23 - 26, 2018 - Halifax, Nova Scotia