Skip to content
Bending the Curve Trial by Trial
?

Testis (male reproductive system)

Testicular germ cell tumors

How platinum chemotherapy, surveillance and risk-adapted care shaped outcomes in testicular germ cell tumors, and what survivors face.

5-year relative survival · United States, SEER 8 registries

79.69%→97.21%

1975 (observed value; first year in the table) to 2018 (observed value; last observed year in the table). Read these with the caveats under “Then and now”.Source

35 milestones, by decade

  • A 1977 report of 50 people with disseminated testicular cancer treated with cisplatin, vinblastine and bleomycin described 74% complete and 26% partial remissions, and an overall 85% disease-free status after surgery for five people with partial remission (abstract; three people were inevaluable because of early death). In a 2002 retrospective, its first author Einhorn wrote that his initial studies with this regimen began in 1974, that metastatic testicular cancer had been associated with only 5% survival in the early 1970s, and that cisplatin-etoposide salvage first used at Indiana University in 1978 was the first time a solid tumour had been cured with second-line chemotherapy. The 1977 series was single-arm and the 2002 statements are the author's own account.
  • A 1969 Nature paper titled 'Platinum compounds: a new class of potent antitumour agents' came from a group that had published in Nature in 1965 on the inhibition of cell division in Escherichia coli by electrolysis products from a platinum electrode; two authors, Rosenberg and VanCamp, appear on both. The FDA's Drugs@FDA data, read through openFDA, list the original cisplatin application (NDA 018057) as approved on 19 December 1978.
  • In US registry data on the SEER Cancer Stat Facts page, observed five-year relative survival for testicular cancer was 79.69% in 1975 and 97.21% in 2018 (SEER 8 series; single-year values, and the page does not say whether the year is the year of diagnosis), while the age-adjusted US death rate was 0.74 per 100,000 men in 1975 and 0.27 in 2024. In the same table the observed 1990 values are already 96.45% survival and 0.28 per 100,000, and the page reports death rates rising on average 2.3% a year over 2015-2024. The page does not say how much of the change reflects treatment, stage, case definition or other factors.

Evidence current to October 2026 · Checked against primary sources to the extent possible

A guided tour of 12 milestones. You can pause or step through it yourself.

Filter milestonesshowing all 35
Treatment type
Setting

Chemotherapy

  1. 1953 · Origin · UnderstandingDixon and Moore's clinicopathological study of testicular tumoursIn 1953 the journal Cancer published a clinicopathological study of testicular tumours by F. J. Dixon and R. A. Moore. A history of gonadal pathology calls it an important paper of the 1950s. This site has not read the paper itself, so it does not describe its findings or any classification it proposed.Dixon FJ and Moore RA, Cancer 1953, "Testicular tumors; a clinicopathological study", reprinted in J Urol in 2002 (PubMed types the reprint as a seminal and historical article). Only bibliographic data were read: PubMed holds no abstract and the text was not opened, so case numbers, histological categories and outcomes are not reported. Young (2005) calls it an important 1950s paper and notes that the authors also wrote the first series fascicle on testicular tumours.
  2. 1972 · Understanding · UnderstandingSkakkebaek reports possible carcinoma in situ of the testisIn 1972 N. E. Skakkebaek published a two-page article in The Lancet titled 'Possible carcinoma-in-situ of the testis'. A later review credits him with first characterising a non-invasive precursor of testicular germ-cell tumours, now called germ cell neoplasia in situ. The article itself was not read for this site.Skakkebaek NE, Lancet 1972: a single-author, two-page article (title only in PubMed; text not read). Later sources describe the lesion: Karellas and Damjanov (2007) credit Skakkebaek with first characterising the non-invasive precursor; Dieckmann and Skakkebaek (1999) report carcinoma in situ next to germ-cell tumour in about 90% of cases, with a 50% probability of progression within 5 years if untreated.
  3. 1975 · Treatment · ChemotherapyBleomycin combinations for testicular cancer (1975)Two 1975 reports described bleomycin combined with other drugs in advanced germ-cell tumours of the testis. In one series, 53 of 70 patients with stage III disease responded, and patients with a complete response lived longer than the others. The other, with continuous-infusion bleomycin and vinblastine, reported complete responses in 9 of 23 patients.Two uncontrolled series. Samuels (Cancer): 83 patients with stage II or III testicular and 7 with extragonadal primaries, bleomycin with vinblastine or a five-drug programme; of 70 stage III, 53 responded (75%): 22 complete, 31 partial; mean survival of complete responders 100+ weeks (3 deaths). Samuels (Cancer Chemother Rep): 23 stage III patients, continuous-infusion bleomycin plus vinblastine; 17 responses, 9 complete (39%); one death from bleomycin pneumonitis.

Molecular origins

  1. 1976 · Detection · UnderstandingSerial hCG and AFP measured to monitor germ-cell tumoursA 1976 report in Cancer followed nine people with germ-cell tumours, measuring two blood proteins, hCG and AFP, repeatedly during treatment. In all nine, one or both proteins helped follow the disease. In two people the tumour masses did not shrink while the markers fell, and surgery found mature teratoma.Perlin E et al., Cancer 1976: case series of nine patients with germinal cell tumours; serial hCG and AFP by double-antibody radioimmunoassay before and during treatment. Eight had an hCG marker and five an AFP marker, four both. Serial values were useful for monitoring disease activity during therapy in all nine. In two, masses did not diminish during chemotherapy while markers fell; surgery showed mature teratoma (one with a microscopic embryonal focus). Primary sites are not stated in the abstract.
  2. 1977 · Treatment · ChemotherapyPVB: high remission rates in disseminated testis cancer (1977)In 1977, 50 patients with disseminated testicular cancer were treated with cisplatin, vinblastine and bleomycin. Of those who could be assessed, about three in four had a complete remission and a quarter a partial remission. In a 2002 review, the lead author wrote that these studies changed the cure rate for disseminated disease from 5% to 60%.Single-arm series: 50 patients with disseminated testicular cancer given PVB (cisplatin, vinblastine, bleomycin); 3 inevaluable (early death). Complete remission 74%, partial remission 26%; 85% disease-free after surgery for residual disease in 5 patients with partial remission. Two drug-related deaths. At 6+ to 30+ months, 38 patients alive and 32 alive and disease-free. Uncontrolled; short follow-up.
  3. 1982 · Understanding · Understandingi(12p) proposed as a specific change in testicular tumoursIn 1982 a short letter in The Lancet asked whether a particular chromosome change, an isochromosome of the short arm of chromosome 12 written i(12p), is specific to testicular tumours. A fuller 1983 report found a small marker chromosome in ten seminomas, one mixed tumour and one malignant teratoma.Atkin NB and Baker MC, Lancet 1982: letter asking whether i(12p) is a specific chromosome change in testicular tumours (title only; text not read). Fuller report, Cancer Genet Cytogenet 1983: a small marker chromosome, often in duplicate, in each of ten seminomas, one combined seminoma and teratoma, and one malignant teratoma; identified as i(12p) in the four most favourable tumours, all seminomas. The authors suggest it may be specific to malignant testicular tumours.
  4. 1982 · Understanding · Surgery⚑ needs reviewDistribution of nodal metastases in non-seminoma (Donohue, 1982)In 1982 Donohue and colleagues published a report on how non-seminoma testis cancer is distributed in the lymph nodes behind the abdomen. The paper has no abstract in the medical literature database, so its findings are not described here.Donohue JP, Zachary JM, Maynard BR, 'Distribution of nodal metastases in nonseminomatous testis cancer', J Urol, August 1982. No abstract in PubMed; the text was not read, so no findings or numbers are recorded. Donohue's 2003 historical review states that RPLND for low-stage disease has been reduced to modified templates and nerve-sparing techniques.
  5. 1987 · Treatment · ChemotherapyAdjuvant chemotherapy vs observation after RPLND, stage IIAfter surgery to remove lymph nodes, some men with testicular cancer still had cancer in the nodes. A randomised study compared two cycles of immediate chemotherapy with careful observation and treatment only at relapse. Chemotherapy almost always prevented relapse, but relapses on observation were usually treated successfully.Williams 1987: 195 evaluable patients with completely resected pathological stage II testicular cancer randomised (1979-1984) after orchiectomy and RPLND to two cycles of cisplatin-based adjuvant chemotherapy (n=97) or observation with treatment at relapse (n=98). Median follow-up 4 years: relapse 6% vs 49%; 94 of 97 survived with adjuvant chemotherapy, and 93 of 98 were alive and disease-free after observation. The authors concluded outcomes were equivalent when surgery, follow-up and chemotherapy are optimal.
  6. 1987 · Treatment · ChemotherapyBEP versus PVB: etoposide replaces vinblastineIn a randomised trial of 261 men with disseminated germ-cell tumours, etoposide replaced vinblastine in a cisplatin and bleomycin regimen. Among 244 evaluable patients, about 83 in 100 on the etoposide regimen and 74 in 100 on the vinblastine regimen became disease-free, a difference that was not statistically significant. Etoposide caused fewer nerve and muscle side effects.Randomised trial, 261 men with disseminated germ-cell tumours: cisplatin and bleomycin with vinblastine (PVB) or etoposide (BEP). Among 244 evaluable, disease-free with or without surgery 74% vs 83% (P not significant); in 157 with high tumour volume 61% vs 77% (P < 0.05). Survival was higher with etoposide (P = 0.048). Fewer paresthesias with etoposide (P = 0.02). Authors concluded the etoposide regimen was superior.
  7. 1988 · Treatment · SurgeryNerve-sparing retroperitoneal lymphadenectomy to preserve ejaculationLymph node surgery in the back of the abdomen can damage nerves that control ejaculation. A 1988 series described identifying and sparing individual nerves during a full bilateral dissection. The technique was feasible in two of three consecutive patients, and nine in ten of those men could still ejaculate.Jewett 1988 (30 consecutive patients, retroperitoneal lymphadenectomy for testis tumour): nerve-sparing bilateral dissection was technically feasible in 20 of 30, and was impractical with extensive gross disease. Of the 20, 18 (90%) ejaculated; the 20 included 8 with bulky (5 cm or more) residual retroperitoneal disease and 12 with nonbulky disease, all of whom ejaculated. No retroperitoneal recurrences with short follow-up.
  8. 1989 · De-escalation · ChemotherapyThree cycles of BEP for good-prognosis diseaseIn a randomised trial of 184 patients with favourable-prognosis disseminated germ-cell tumours, three courses of cisplatin, etoposide and bleomycin gave results similar to four: about nine in ten patients in each group were continuously disease-free. The authors preferred three courses, and 10-year follow-up from one centre showed no significant survival difference.Southeastern Cancer Study Group randomised phase III, 184 patients with favourable-prognosis (minimal or moderate extent) disseminated germ cell tumours: 3 vs 4 courses of BEP. Initial disease-free status 98% (86/88) vs 97% (93/96); continuously disease-free 92% vs 92% (81/88; 88/96); 10 relapses, 5 per arm. Indiana follow-up (118 patients, median 10.1 years): OS P = .80, DFS P = .93. Authors preferred three courses.
  9. 1992 · De-escalation · SurgerySurveillance after orchidectomy for stage I non-seminomaA UK-led study followed men with early-stage non-seminoma testicular cancer after surgery alone, with treatment only if the cancer returned. About three in four stayed free of relapse at two years and nearly all were alive at five years. Men with several risky tumour features relapsed more often.MRC prospective surveillance study (stage I NSGCT, orchidectomy alone): 373 of 396 patients entered 1984-1987 were eligible. Relapse-free rate 75% (95% CI 71-79%) at 2 years and 73% at 5 years; 5-year survival 98%; relapse-free rate 54% with three or four risk factors. Randomised TE08 (two vs five surveillance CT scans, median follow-up 40 months): relapses 15% vs 20%; intermediate-prognosis relapse 0.8% vs 0.6%, difference 0.2% (90% CI -1.2 to 1.6).
  10. 1995 · Reversal · ChemotherapyIs bleomycin needed? EP versus BEP in good-prognosis diseaseThree randomised trials tested leaving bleomycin out of first-line chemotherapy for good-prognosis non-seminoma. Two found worse results without it: 95% versus 87% had a complete response in one, and lower failure-free survival in the other. A French trial of three cycles with bleomycin versus four without found similar results but was underpowered.ECOG (178 randomised; 3 cycles): disease-free 94% BEP vs 88% EP; failure-free 86% vs 69% (P = .01); OS 95% vs 86% (P = .01). EORTC (419; 4 cycles): complete response 95% BEP vs 87% EP (P = .0075); relapses 4% per arm; survival P = .262; 2 bleomycin pulmonary deaths on BEP. GETUG T93BP (257; 3 BEP vs 4 EP): 4-year EFS 91% vs 86% (P = 0.135); underpowered. Doses differ between trials.
  11. 1996 · Treatment · ChemotherapyAdjuvant BEP for high-risk stage I non-seminomaSome men with early-stage non-seminoma have a high chance of relapse after surgery. A UK study gave two courses of BEP chemotherapy to more than a hundred high-risk men and saw two relapses. Later work tested one course: a German trial against lymph node surgery, and a Swedish and Norwegian population-based programme.MRC single-arm study: 114 eligible high-risk patients given two courses of BEP, two relapses at median follow-up 4 years. AUO AH 01/94 (randomised, 382 patients): 2-year recurrence-free survival 99.46% with one course of BEP vs 91.87% with RPLND (HR for recurrence with surgery 7.937, 95% CI 1.808-34.48, P=0.0011). SWENOTECA (non-randomised): 5-year relapse after one course of BEP 3.2% with lymphovascular invasion, 1.6% without.
  12. 1997 · Reversal · ChemotherapyCarboplatin in place of cisplatin was inferior at trial dosesIn a trial of 598 patients with good-risk non-seminoma, four cycles of BEP were compared with the same regimen using carboplatin instead of cisplatin. Complete response was 94% with cisplatin and 87% with carboplatin, and three-year survival was 97% versus 90%. The authors concluded the carboplatin regimen was inferior at the doses used.MRC/EORTC randomised trial, 598 patients with good-risk NSGCT, 4 cycles of BEP vs CEB (etoposide 120 mg/m2 days 1-3, bleomycin 30 U). Complete response 94.4% (253/268) vs 87.3% (227/260), P = .009. Failure-free at 1 year 91% (95% CI 88-94) vs 77% (72-82), P < .001. 3-year survival 97% (95-99) vs 90% (86-94), P = .003 (10 vs 27 deaths). Carboplatin regimen inferior at these doses.
  13. 1997 · Understanding · UnderstandingIGCCCG prognostic classification of metastatic germ-cell cancerCollaborative groups from ten countries pooled data on 5,202 people with non-seminoma and 660 with seminoma treated with cisplatin-containing chemotherapy. They defined good, intermediate and poor prognosis groups, with five-year survival of about 91, 79 and 48 in 100.Pooled data from collaborative groups in 10 countries: 5,202 NSGCT and 660 seminoma treated with cisplatin-containing chemotherapy; median follow-up 5 years. NSGCT adverse factors: mediastinal primary, raised AFP, hCG and LDH, non-pulmonary visceral metastases; seminoma: non-pulmonary visceral metastases. 5-year survival: good prognosis (60% of germ cell tumours) 91% (89-93); intermediate (26%) 79% (75-83); poor (14%, non-seminoma only) 48% (42-54). Validated on an independent data set.

Targeted therapy

  1. 1998 · Treatment · ChemotherapyIfosfamide-based salvage: VeIP, then TIPTwo series, and a later update, described ifosfamide-based chemotherapy for germ-cell tumours that progressed after cisplatin. With VeIP in 135 patients, about half became disease-free, and one in four stayed free of disease. With TIP in 46 selected favourable-prognosis patients, 63% had a durable complete response; a 2024 report of 104 patients gave 5-year PFS 66%.Single-arm series. VeIP second line (135 patients, 1984-1989): disease-free 49.6% after chemotherapy with or without surgery; continuously disease-free 23.7%. TIP second line (46 patients with testicular primary and prior complete response): complete response 70%, durable complete response 63%, 2-year PFS 65% (95% CI 51-79), median follow-up 69 months. TIP long-term (104 patients, median 8.9 years): 5-year PFS 66% (55-74), OS 69% (59-77).
  2. 1999 · De-escalation · Radiation⚑ needs reviewSmaller radiotherapy fields and lower doses for stage I seminomaAfter surgery for stage I seminoma, radiotherapy was given to abdominal lymph nodes. Two MRC trials tested a smaller field and a lower dose. Relapse rates were low in all groups, the smaller field caused less short-term toxicity, and the lower dose was judged unlikely to raise relapse by more than a few percentage points.MRC TE10 (478 men, 30 Gy): para-aortic strip vs dogleg field, 3-year relapse-free survival 96% vs 96.6%; 18 relapses (nine in each group), four pelvic, all after para-aortic radiotherapy. TE18/EORTC 30942 (625 men): 30 vs 20 Gy, median follow-up 61 months, 10 vs 11 relapses (HR 1.11, 90% CI 0.54-2.28); lethargy at 4 weeks 20% vs 5%. Later combined report (TE18 counted as 1094 patients): 20 Gy HR 0.63 (90% CI 0.38-1.04).
  3. 2001 · Detection · UnderstandingFDG-PET to predict viable tumour in residual seminoma massesAfter chemotherapy for seminoma, scans often show leftover masses, and many contain no living tumour. A prospective multicentre study found PET scanning predicted living tumour in these masses more reliably than the size of the mass. A small later study found false-positive PET results to be common.SEMPET (prospective, multicentre; 2004 update, 51 patients, 56 scans): FDG-PET correctly predicted all 19 residual lesions above 3 cm and 35 (95%) of 37 lesions of 3 cm or less; specificity 100% (95% CI 92-100%), sensitivity 80% (44-95%), PPV 100%, NPV 96%, versus 74%, 70%, 37% and 92% for the CT size criterion. A separate 20-patient study: PPV 0.25 and NPV 1, with 9 false-positive PET scans.
  4. 2002 · De-escalation · Understanding⚑ needs reviewSurveillance and risk-adapted management of stage I seminomaAfter surgery for stage I seminoma, most men do not relapse. A 2002 pooled analysis of surveillance series found larger tumours and invasion of the rete testis predicted relapse. Later programmes used these risk factors to offer carboplatin or surveillance, and a 2024 analysis proposed three risk groups.Warde 2002 pooled analysis (638 patients, four surveillance series, median follow-up 7.0 years): 5-year relapse-free rate 82.3%; tumour above 4 cm HR 2.0 (95% CI 1.3-3.2), rete testis invasion HR 1.7 (1.1-2.6). SWENOTECA (non-randomised, 897 patients): relapse 4.0% on surveillance and 2.2% after carboplatin without risk factors; 15.5% and 9.3% with one or two. EAU analysis (1,016 patients): 5-year relapse 8%, 20% and 44% by risk group.
  5. 2003 · Understanding · Supportive care⚑ needs reviewCardiovascular disease as a late effect of testicular cancer treatmentFrom 2003, cohorts followed for many years after testicular cancer reported more heart and blood-vessel disease, or related deaths, after radiotherapy or cisplatin-based chemotherapy, compared with surgery alone or the general population. Estimates differed between studies and not every one reached significance. A 2026 analysis tested a risk-prediction tool in men treated with contemporary chemotherapy.Huddart 2003 (UK, 992 patients, median follow-up 10.2 years): age-adjusted relative risk of cardiac events 2.59 after chemotherapy, 2.40 after radiotherapy, 2.78 after both (reference group not stated in the abstract). Van den Belt-Dusebout 2006 (2,512 five-year survivors): BEP, myocardial infarction HR 1.2 (95% CI 0.7 to 2.1), not significant. Haugnes 2010 (990 men): atherosclerotic disease HR 2.6 (1.1 to 5.9), chemotherapy versus surgery only. Observational; no pooling.
  6. 2003 · Understanding · SurgerySurgery for small residual retroperitoneal masses after chemotherapyAfter chemotherapy for non-seminoma, a small mass can remain behind the abdomen. In 87 men whose masses were 20 millimetres or smaller, about one in three still contained living tumour or teratoma, and no scan or clinical measure predicted which. The authors concluded surgery remained necessary.Oldenburg 2003 (87 patients with metastatic NSGCT, residual retroperitoneal mass 20 mm or less on CT after cisplatin- or carboplatin-based chemotherapy, RPLND): complete fibrosis or necrosis 58 (67%), teratoma 23 (26%), vital malignant germ-cell tumour 6 (7%, one rhabdomyosarcoma). In five of the six, the lesion was 10 mm or less. No pre- or postchemotherapy clinical or radiologic parameter predicted histology.
  7. 2005 · Understanding · Supportive careSecond cancers among 40,576 testicular cancer patientsIn 2005 a study combining 14 population cancer registries followed more than 40,000 men with testicular cancer and found a raised risk of second solid cancers that lasted at least 35 years. Later cohorts examined specific treatments, and a 2026 review linked chemotherapy to a higher relative risk of leukaemia, though absolute risk stayed small.Travis 2005: 40,576 one-year survivors in 14 registries (1943 to 2001), 2,285 second solid cancers. In 10-year survivors diagnosed at age 35, relative risk 1.9 (95% CI 1.8 to 2.1), still raised at 35 years (1.7, 1.5 to 2.0). By treatment, chemotherapy alone: RR 1.8 (1.3 to 2.5). Fung 2013 (SEER non-seminoma): SIR 1.43 (1.18 to 1.73) after chemotherapy, 0.93 after surgery. Registry cohorts, not randomised.
  8. 2005 · De-escalation · ChemotherapyTE19: single-dose carboplatin vs radiotherapy, stage I seminomaAfter surgery for stage I seminoma, radiotherapy to the abdominal lymph nodes was a usual next step. A large randomised trial compared it with one carboplatin infusion. Relapse-free rates were similar, and fewer new tumours in the other testis were reported with carboplatin. The authors concluded that either approach could be used.MRC TE19/EORTC 30982: 1,477 patients randomised to radiotherapy (n=904) or one carboplatin dose (n=573). Relapse-free rate 96.7% vs 97.7% at 2 years and 95.9% vs 94.8% at 3 years (HR 1.28, 90% CI 0.85-1.93, p=0.32; non-inferiority concluded). At median 6.5 years, 5-year rate 96.0% vs 94.7% (HR 1.25, 90% CI 0.83-1.89); contralateral germ-cell tumours 15 vs 2 (HR 0.22 for carboplatin, 95% CI 0.05-0.95, p=0.03).
  9. 2007 · Reversal · Chemotherapy⚑ needs reviewFirst-line high-dose chemotherapy did not show benefit in trialsTwo randomised trials added high-dose chemotherapy with stem-cell rescue to first-line treatment of poor-prognosis germ-cell tumours. In one of 219 patients (intermediate or poor risk), 52% with high-dose chemotherapy and 48% with BEP alone had a durable complete response at one year, not significantly different. A 131-patient European trial, closed early, also found no significant gain.Motzer 2007: phase 3, 219 patients with intermediate or poor-risk germ cell tumours, BEP x4 vs BEP x2 plus high-dose carboplatin-containing chemotherapy: 1-year durable complete response 52% vs 48% (P = .53); in 67 patients with unsatisfactory marker decline 61% vs 34% (P = .03). EORTC 30974 (137 randomised, closed early; 131 analysed): complete response 44.6% vs 33.3% (P = 0.18); 2-year FFS 58.2% vs 44.8% (P = 0.060), high-dose VIP vs BEP.
  10. 2007 · Treatment · Chemotherapy⚑ needs reviewHigh-dose chemotherapy with stem-cell rescue for relapsed diseaseA 2007 retrospective review of 184 patients whose testicular cancer progressed after cisplatin-based chemotherapy found 116 in complete remission without relapse after high-dose carboplatin and etoposide with stem-cell rescue. It was not a randomised trial. An earlier randomised trial of one high-dose course found no significant gain.Einhorn 2007: retrospective review (not randomised), 184 consecutive patients progressing after cisplatin chemotherapy; 173 had two courses of carboplatin 700 mg/m2 plus etoposide 750 mg/m2 with stem-cell rescue. 116 in complete remission without relapse (median follow-up 48 months); 94 of 135 second-line; 22 of 49 third-line or later; 3 drug-related deaths. IT-94 (Pico 2005, randomised, 280): one high-dose course, 3-year EFS 35% vs 42% (P = 0.16).
  11. 2009 · Understanding · UnderstandingGenome-wide studies link KITLG and loci to testicular GCTIn 2009 two research groups published genome-wide scans comparing the DNA of people with testicular germ cell tumours and controls. Both found common inherited variants linked to higher risk, including variants in or near KITLG, a gene for the ligand of the KIT receptor, and a region on chromosome 5.Two case-control genome-wide association studies, Nat Genet 2009. Rapley et al. (UK discovery 730 cases, 1,435 controls; replication 571 cases, 1,806 controls): per-allele OR 2.55 (95% CI 2.05-3.19) on chromosome 12, where KITLG may explain the signal; 1.37 (1.19-1.58) on chromosome 5; 1.50 (1.28-1.75) on chromosome 6. Kanetsky et al. (discovery 277 cases, 919 controls; replication ORs): KITLG 3.08 (2.29-4.13); 5q31.3 1.37 (1.14-1.64).

Immunotherapy

  1. 2014 · Treatment · ChemotherapyGETUG 13: tumour-marker decline guides dose-dense chemotherapyAfter one cycle of BEP in poor-prognosis non-seminoma, patients whose tumour markers fell too slowly were randomised to continue BEP or switch to a dose-dense regimen. Three-year progression-free survival was about 59 in 100 versus 48 in 100; at a median of seven years, five-year overall survival was 71% versus 61%, not statistically significant.GETUG 13, phase 3, 263 patients with IGCCCG poor-prognosis NSGCT; after 1 cycle of BEP, 203 with unfavourable marker decline randomised to BEP or dose-dense chemotherapy (paclitaxel-BEP-oxaliplatin, then cisplatin-ifosfamide-bleomycin). 3-year PFS 59% vs 48% (HR 0.66; 95% CI 0.44-1.00; P = 0.05). Update, median 7.1 years: 5-year PFS 58.9% vs 46.7% (HR 0.65; 0.44-0.97; P = 0.036); OS 70.9% vs 61.3% (HR 0.74; 0.46-1.20; P = 0.22).
  2. 2018 · Understanding · Supportive care⚑ needs reviewCumulative burden of morbidity after cisplatin-based chemotherapyA 2018 multi-centre study of more than 1,200 men who had finished standard cisplatin-based chemotherapy for testicular cancer scored their combined load of long-term health problems. About one in five had a high or very high score, and some problems, such as hearing loss and tinnitus, tended to occur together.Kerns 2018: 1,214 men aged 55 or younger at diagnosis, at least 1 year after first-line chemotherapy. Cumulative burden of morbidity score: low or very low 47%, medium 30%, high 15%, very high or severe 4.1% (as published; sum 96.1%). Higher scores were associated with four cycles of BEP (OR 1.44, 95% CI 1.04 to 1.98) or IEP (OR 1.96, 1.04 to 3.71), reference group not stated in the abstract; score did not differ between regimens (P = .36).
  3. 2019 · Detection · UnderstandingProspective study of serum miR-371a-3p in testicular GCTIn 2019 a multicentre study measured a small RNA molecule, microRNA-371a-3p, in the blood of 616 men with testicular germ cell tumours and 258 male controls. The test outperformed the established blood markers, and the marker was expressed in all tumour types except teratoma. The authors called for further validation before clinical implementation.Dieckmann KP et al., J Clin Oncol 2019: prospective multicentre study, 616 patients with testicular GCT (359 seminoma, 257 nonseminoma) and 258 male controls; serum miR-371a-3p by qPCR (M371 test). Primary diagnosis, preoperative samples from 522 patients vs controls: sensitivity 90.1%, specificity 94.0%, AUC 0.966; PPV 97.2% (four institutions). AFP, β-hCG and LDH sensitivities were below 50% in seminoma. Teratoma did not express the marker. Authors: consider after further validation.

Precision & engineered

  1. 2021 · Understanding · UnderstandingIGCCCG Update: modern survival estimates and refined prognosisTwo 2021 papers re-examined the IGCCCG groups in men treated from 1990 to 2013: 9,728 with non-seminoma and 2,451 with seminoma. Five-year survival was about 96, 89 and 67 in 100 for good, intermediate and poor non-seminoma, and 95 and 88 in 100 for good and intermediate seminoma.Pooled trial and registry data, cisplatin- and etoposide-based first-line chemotherapy, 1990-2013. NSGCT (n=9,728): 5-year OS 96% good, 89% intermediate, 67% poor; PFS 90%, 78%, 54%. Seminoma (n=2,451): 5-year OS 95% good, 88% intermediate; PFS 89%, 79%. New factors: NSGCT LDH cut-off 2.5 x ULN, age, lung metastases; seminoma LDH above 2.5 x ULN (good prognosis 3-year PFS 80% vs 92%).
  2. 2021 · Understanding · Supportive careNon-testicular-cancer mortality in long-term survivors (cisplatin era)A 2021 Norwegian registry study followed men diagnosed with testicular cancer from 1980 to 2009 and found more deaths from other causes than expected after platinum-based chemotherapy or radiotherapy, led by second cancers. Beyond ten years, death risk exceeded that after surgery at four or more cycles. A smaller cohort linked lifestyle and psychosocial problems with higher mortality.Hellesnes 2021: 5,707 Norwegian men diagnosed 1980 to 2009, median follow-up 18.7 years; 665 (12%) died of causes other than testicular cancer. SMR for non-testicular-cancer death 1.23 (95% CI 1.14 to 1.33) overall; 1.23 after platinum-based chemotherapy, 1.28 after radiotherapy; second cancers 1.53. Versus surgery beyond 10 years, more than four cycles: HR 2.04 (95% CI 1.25 to 3.35). Fosså 2022 (775 men): cumulative 20-year mortality 14%.
  3. 2022 · De-escalation · SurgeryPrimary RPLND for stage IIA/B seminoma: PRIMETESTStage IIA or IIB seminoma is usually treated with radiotherapy or chemotherapy. In this small single-arm trial, 33 men had lymph node surgery alone. About one in three had a recurrence, which missed the trial's target, so the authors did not recommend surgery alone outside trials.PRIMETEST (single-arm, single-centre phase 2; 33 patients, 2016-2021) of primary RPLND alone for seminoma, clinical stage IIA/B. After median follow-up 32 months, ten recurrences (30%, 95% CI 16-49%); primary endpoint (progression-free survival at 36 months, positive if fewer than 30% recurred) not met. Three of ten were infield recurrences; 3 of 33 (9%) had pN0 disease. Authors: surgery alone not recommended outside a trial.
  4. 2022 · De-escalation · UnderstandingTRISST: fewer scans and MRI instead of CT in seminoma surveillanceMen under surveillance for stage I seminoma have regular scans. The TRISST trial randomised them to three or seven scans, and to CT or MRI. Advanced relapse was rare in every group. MRI and the shorter schedule were statistically non-inferior, although the shorter schedule had slightly more advanced relapses.TRISST (phase III non-inferiority, 2x2 factorial; 669 patients, 35 UK centres, median follow-up 72 months): 6-year incidence of stage IIC or higher relapse 2.8% with three scans vs 0.3% with seven (absolute increase 2.5%, 90% CI 1.0-4.1; non-inferior), and 0.6% with MRI vs 2.6% with CT (1.9% decrease; non-inferior). Ten such events overall; 82 patients (12%) relapsed; 5-year survival 99%, no tumour-related deaths.
  5. 2026 · Today · Chemotherapy⚑ needs reviewEAU 2026 guideline: current chemotherapy recommendationsThe 2026 European Association of Urology guideline text sets out first-line chemotherapy by IGCCCG group: three cycles of BEP for good-prognosis non-seminoma, four for intermediate prognosis, and for poor prognosis four cycles of BEP with intensification considered after an unfavourable marker decline. For first relapse it lists four cycles of VIP, TIP or GIP.EAU Guidelines on Testicular Cancer, 2026 edition (web text read October 2026). Good-risk non-seminoma: BEP x3 standard; EP x4 only with a clear contraindication to bleomycin; omission otherwise not recommended. Intermediate: BEP x4. Poor prognosis: BEP x4; primary high-dose chemotherapy has shown no overall survival benefit in RCTs; dose-dense switch after unfavourable marker decline described. First relapse: four cycles of VIP, TIP or GIP (no RCT compares them); high-dose versus conventional dose is being tested in TIGER.

Then and now

  • 5-year relative survival · United States, SEER 8 registries

    • 79.69%

      1975 (observed value; first year in the table)

    Males, all races (chart note for the trend table: 'All Races, Males'); testicular cancer; the page states no age range or histology definition.

    Observed single-year value; the adjacent modelled-trend value for 1975 is 82.28% and the observed values for 1976 and 1977 are 82.98% and 86.12%. SEER 8 five-year relative survival table on the SEER Cancer Stat Facts testicular page. The table column is headed 'Year' and the chart is titled 'SEER 8 5-Year Relative Survival Percent from 1975-2018'; neither labels the year as year of diagnosis. Single-year values fluctuate. The page shows no stage-specific survival and does not attribute any change to a treatment or other cause. Compare only with other values from the same SEER 8 series.

    Source

  • 5-year relative survival · United States, SEER 8 registries

    • 96.45%

      1990 (observed value)

    Males, all races (chart note for the trend table: 'All Races, Males'); testicular cancer; the page states no age range or histology definition.

    Observed single-year value; modelled trend for 1990 is 95.96%. Included to show when in the series the level reached its present range: the modelled trend is 94.84% in 1986 and 95.91% in 1988, and stays between 95.9% and 96.8% in every later year of the table. SEER 8 five-year relative survival table on the SEER Cancer Stat Facts testicular page. The table column is headed 'Year' and the chart is titled 'SEER 8 5-Year Relative Survival Percent from 1975-2018'; neither labels the year as year of diagnosis. Single-year values fluctuate. The page shows no stage-specific survival and does not attribute any change to a treatment or other cause. Compare only with other values from the same SEER 8 series.

    Source

  • 5-year relative survival · United States, SEER 8 registries

    • 97.21%

      2018 (observed value; last observed year in the table)

    Males, all races (chart note for the trend table: 'All Races, Males'); testicular cancer; the page states no age range or histology definition.

    Observed single-year value; modelled trend for 2018 is 96.64%, and neighbouring observed years range from 94.98% (2017) to 98.44% (2006). Use the modelled trend or the multi-year SEER 21 figure for a more stable estimate. SEER 8 five-year relative survival table on the SEER Cancer Stat Facts testicular page. The table column is headed 'Year' and the chart is titled 'SEER 8 5-Year Relative Survival Percent from 1975-2018'; neither labels the year as year of diagnosis. Single-year values fluctuate. The page shows no stage-specific survival and does not attribute any change to a treatment or other cause. Compare only with other values from the same SEER 8 series.

    Source

  • 5-year relative survival · United States, SEER 21 registries (excluding IL)

    • 94.6%

      2016-2022 (as labelled on the page)

    Testicular cancer; the page does not state age, race or sex for this headline figure (the page and the trend-table notes cover males, all races).

    Different registry grouping from the SEER 8 series above and a multi-year figure, so it is not directly comparable with the 1975 value and should not be subtracted from the SEER 8 values. The SEER 8 single-year values for 2016 and 2018 on the same page are higher (96.62% and 97.21%). The page states that relative survival excludes the risk of dying from other causes and that these statistics cannot predict what will happen to an individual. The page shows no stage-specific survival.

    Source

  • Age-standardised mortality · United States (national mortality)

    • 0.74per 100,000

      1975 (observed value; first year in the table)

    Males, all races; age-adjusted rate.

    The adjacent modelled-trend value for 1975 is 0.78. US mortality (National Center for Health Statistics), not SEER registry data; the page states that deaths come from U.S. Mortality. Single-year observed value. Mortality is shown beside survival because survival can rise through case-mix and detection changes as well as treatment; the page does not separate these. Cause-of-death coding and the standard population used for age adjustment over five decades are not described on the page.

    Source

  • Age-standardised mortality · United States (national mortality)

    • 0.28per 100,000

      1990 (observed value)

    Males, all races; age-adjusted rate.

    Modelled trend for 1990 is 0.30. US mortality (National Center for Health Statistics), not SEER registry data; the page states that deaths come from U.S. Mortality. Single-year observed value. Mortality is shown beside survival because survival can rise through case-mix and detection changes as well as treatment; the page does not separate these.

    Source

  • Age-standardised mortality · United States (national mortality)

    • 0.27per 100,000

      2024 (observed value; last year in the table)

    Males, all races; age-adjusted rate.

    The 2023 observed value is 0.32 and the 2020-2024 pooled rate on the page is 0.3 per 100,000 men per year. The lowest observed values in the table are 0.23 (2007 and 2015). The page states that age-adjusted death rates have been rising on average 2.3% each year over 2015-2024 (and new-case rates 0.7% each year over 2014-2023). US mortality (National Center for Health Statistics), not SEER registry data; the page states that deaths come from U.S. Mortality. Single-year observed value. Mortality is shown beside survival because survival can rise through case-mix and detection changes as well as treatment; the page does not separate these.

    Source

  • 10-year relative survival · United Kingdom

    • 62.5%

      Diagnosed in the 1970s

    Males diagnosed with testicular cancer (the page's wording); no age range stated for this figure.

    Cancer Research UK words this as 'survived their disease beyond ten years'. The page does not say whether the figure is net or relative survival, which UK nations are included for the 1970s, or how patients diagnosed then were followed. It says that survival trends reflect a combination of changes in treatment and stage distribution. Different source, country and measure from the SEER series; do not compare directly with them.

    Source

  • 10-year relative survival · United Kingdom

    • 96.5%

      2018 (the page says 'predicted')

    Males diagnosed with testicular cancer in the UK; the page also gives 97.6% for ages 15-44 and 56.9% for ages 75-99 (UK, 2018).

    The page states that it is predicted that more than 9 in 10 (96.5%) survive ten years or more (2018); this is a predicted figure, not observed ten-year follow-up of that year's cases, and the page does not describe the prediction method or whether it is net or relative survival. The trend sentence on the same page gives 96.5% for 2018 without the word 'predicted'. Different source, country and measure from the SEER series.

    Source

Survival figures are affected by earlier detection (lead time) and overdiagnosis, and by changes in who is diagnosed. For screened cancers, read them alongside mortality.

Evidence chains

A lane follows one defined population and setting, where each new trial tests against what was then standard. Compare only within a lane, and read every comparison as illustrative.

Stage I seminoma, after orchidectomy

Men with clinical stage I seminoma after orchidectomy, treated with adjuvant radiotherapy or carboplatin or followed by surveillance · Adjuvant · endpoint: RFR

Results are not comparable across trials.

The trials in this lane ask different questions: TE10 compared radiotherapy fields (para-aortic strip against dogleg) and TE19 compared single-dose carboplatin with radiotherapy, so the control arm is a different standard in each. TE10's value is a 3-year relapse-free rate (called relapse-free survival in the abstract) and TE19's chart value is the 2-year rate, so only TE10 shares the plotted axis. TE18 (20 Gy against 30 Gy) reports a hazard ratio of relapse without per-arm rates and its patient count is unresolved between reports (needs review), so it is not plotted. The surveillance milestone describes pooled and population-based series, not randomised arms. Follow-up, trial era and per-protocol or intention-to-treat analysis differ; compare only each trial's own two arms.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

020406080100RFR at 36 months (%)1999MRC TE10
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
RFR at 36 months, rate, across 1 trial in this lane. MRC TE10 (1999): experimental 96 % vs control 96.6 %.
Data table
TrialYearEndpointControl (%)Experimental (%)
MRC TE101999RFR96.6 %96 %
MRC TE19 / EORTC 309822005RFR (not plotted: different measure or time point)96.7 %97.7 %

Stage I non-seminoma, adjuvant chemotherapy

Men with clinical stage I non-seminomatous germ-cell tumour after orchidectomy, with adjuvant BEP compared against retroperitoneal lymph node dissection or against surveillance · Adjuvant · endpoint: RFS

Results are not comparable across trials.

Only the German AUO AH 01/94 trial randomised one course of BEP against lymph node dissection (RPLND), and it reports a 2-year recurrence-free rate. The UK single-arm study and the Swedish-Norwegian programme (non-randomised, risk-adapted) have no randomised control arm and are not plotted. Surveillance results from the 1990s used different risk definitions and are described in a separate milestone. Compare only the two arms of the AUO trial.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

020406080100RFS at 24 months (%)1996AUO AH 01/94
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
RFS at 24 months, rate, across 1 trial in this lane. AUO AH 01/94 (1996): experimental 99.46 % vs control 91.87 %.
Data table
TrialYearEndpointControl (%)Experimental (%)
AUO AH 01/941996RFS91.87 %99.46 %

Disseminated germ-cell tumours, first-line cisplatin regimens

Men with disseminated or metastatic germ-cell tumours, mostly non-seminoma and mostly good prognosis in the randomised trials, receiving first-line cisplatin-based combination chemotherapy · Metastatic · endpoint: FFS

Results are not comparable across trials.

Endpoints differ. Only the ECOG and MRC/EORTC results are failure-free survival rates (ECOG without a stated time point, MRC/EORTC at 1 year). The 1987 BEP against PVB trial and the 1989 three against four cycles trial report disease-free status after treatment (with or without surgery), a response measure, and the GETUG T93BP trial reports 4-year event-free survival; these are drawn with a distinct marker, and results at a different time point from the first plotted entry are listed rather than plotted. The 1977 PVB report is a single-arm series with no comparator and is not plotted. Risk definitions (Indiana staging in the 1980s trials, IGCCCG-based later), cycle numbers, etoposide doses, and follow-up differ, and several trials were underpowered for survival. Compare only each trial's own two arms.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

020406080100FFS (%)198719891995williams-1987-bep-vs-pvbsectg-bep-3-vs-4-cyclesecog-1995-bleomycin-omission
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
FFS, rate, across 3 trials in this lane. williams-1987-bep-vs-pvb (1987): experimental 83 % vs control 74 %; sectg-bep-3-vs-4-cycles (1989): experimental 92 % vs control 92 %; ecog-1995-bleomycin-omission (1995): experimental 69 % vs control 86 %.
Data table
TrialYearEndpointControl (%)Experimental (%)
williams-1987-bep-vs-pvb1987Disease-free status after treatment (differs from lane)74 %83 %
sectg-bep-3-vs-4-cycles1989Continuous disease-free rate (differs from lane)92 %92 %
ecog-1995-bleomycin-omission1995FFS86 %69 %
GETUG T93BP1995EFS (differs from lane) (not plotted: different measure or time point)91 %86 %
mrc-eortc-1997-bep-vs-ceb1997FFS (not plotted: different measure or time point)91 %77 %

Intermediate and poor prognosis, first-line intensification

Previously untreated metastatic germ-cell tumours of intermediate or poor prognosis; the GETUG 13 randomisation applies only to poor-prognosis non-seminoma with an unfavourable tumour-marker decline after one cycle of BEP · Metastatic · endpoint: PFS

Results are not comparable across trials.

Endpoints differ: GETUG 13 reports 3-year progression-free survival (the lane endpoint), EORTC 30974 reports 2-year failure-free survival (closed early at 137 of 222 planned patients; needs review because the published rates are inconsistent between sections), and the Motzer trial reports the rate of durable complete response at 1 year. Entries other than GETUG 13 are therefore drawn with a distinct marker, and any with a different time point or measure from the first plotted entry are listed rather than plotted. The experimental treatments (high-dose chemotherapy with stem-cell rescue, or a dose-dense regimen after a marker-guided switch) and the control (four BEP cycles, or BEP continued) differ, as do risk definitions and era. Compare only each trial's own two arms.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

020406080100PFS at 24 months (%)2007EORTC 30974
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
PFS at 24 months, rate, across 1 trial in this lane. EORTC 30974 (2007): experimental 58.2 % vs control 44.8 %.
Data table
TrialYearEndpointControl (%)Experimental (%)
EORTC 309742007FFS (differs from lane)44.8 %58.2 %
motzer-2007-hdct-first-line2007Durable complete response rate (differs from lane) (not plotted: different measure or time point)48 %52 %
GETUG 132014PFS (not plotted: different measure or time point)48 %59 %

Relapsed or refractory germ-cell tumours, salvage chemotherapy

Men whose germ-cell tumour progressed or relapsed after first-line cisplatin-based chemotherapy · Relapsed / refractory · endpoint: EFS

Results are not comparable across trials.

The ifosfamide-based VeIP and TIP reports are single-arm series, and the Indiana high-dose review is retrospective; none can be drawn as a control against an experimental arm. The IT-94 randomised trial (conventional salvage against conventional cycles then high-dose carboplatin, etoposide and cyclophosphamide) is the only pair, and the assignment of its 3-year event-free survival values to arms follows the abstract's order (needs review). Patient selection, prior therapy and favourable-risk eligibility differ greatly between the series; compare only IT-94's own two arms.

Each pair shows one trial: its control arm and its experimental arm. Yesterday's experimental arm is often today's control. Populations, definitions and follow-up differ.

020406080100EFS at 36 months (%)2007IT-94
  • experimental arm
  • control arm
  • different endpoint from the lane's
  • step line through experimental values
EFS at 36 months, rate, across 1 trial in this lane. IT-94 (2007): experimental 42 % vs control 35 %.
Data table
TrialYearEndpointControl (%)Experimental (%)
IT-942007EFS35 %42 %

Standard of care, then and now

How the usual sequence of treatment changed for one setting. Choose a point in time; each step links to the milestone behind it.

Disseminated germ-cell tumours: first-line chemotherapy

Men with disseminated or metastatic germ-cell tumours, mostly non-seminoma · Metastatic

  1. Chemotherapy1Poor prognosis: intensify if slow declineGETUG 13: tumour-marker decline guides dose-dense chemotherapy (2014)
  2. Chemotherapy2First relapse: VIP, TIP or GIP (EAU 2026)EAU 2026 guideline: current chemotherapy recommendations (2026)
Today. Based only on the trials and guideline text cited in each step. The 2026 EAU text lists regimens by IGCCCG group; GETUG 13 tested marker-guided intensification.
Text version
  • 1977: 1st line: cisplatin, vinblastine, bleomycin
  • 1987: 1st line: cisplatin, etoposide, bleomycin
  • 2000: Good prognosis: BEP, 3 cycles studied
  • Today: Poor prognosis: intensify if slow decline → First relapse: VIP, TIP or GIP (EAU 2026)

Reversals and dead ends

Approaches that were abandoned or withdrawn, and trials whose results were later questioned, are part of the story.

  • 1995

    Is bleomycin needed? EP versus BEP in good-prognosis disease

    Three randomised trials tested leaving bleomycin out of first-line chemotherapy for good-prognosis non-seminoma. Two found worse results without it: 95% versus 87% had a complete response in one, and lower failure-free survival in the other. A French trial of three cycles with bleomycin versus four without found similar results but was underpowered.

    ECOG (178 randomised; 3 cycles): disease-free 94% BEP vs 88% EP; failure-free 86% vs 69% (P = .01); OS 95% vs 86% (P = .01). EORTC (419; 4 cycles): complete response 95% BEP vs 87% EP (P = .0075); relapses 4% per arm; survival P = .262; 2 bleomycin pulmonary deaths on BEP. GETUG T93BP (257; 3 BEP vs 4 EP): 4-year EFS 91% vs 86% (P = 0.135); underpowered. Doses differ between trials.

  • 1997

    Carboplatin in place of cisplatin was inferior at trial doses

    In a trial of 598 patients with good-risk non-seminoma, four cycles of BEP were compared with the same regimen using carboplatin instead of cisplatin. Complete response was 94% with cisplatin and 87% with carboplatin, and three-year survival was 97% versus 90%. The authors concluded the carboplatin regimen was inferior at the doses used.

    MRC/EORTC randomised trial, 598 patients with good-risk NSGCT, 4 cycles of BEP vs CEB (etoposide 120 mg/m2 days 1-3, bleomycin 30 U). Complete response 94.4% (253/268) vs 87.3% (227/260), P = .009. Failure-free at 1 year 91% (95% CI 88-94) vs 77% (72-82), P < .001. 3-year survival 97% (95-99) vs 90% (86-94), P = .003 (10 vs 27 deaths). Carboplatin regimen inferior at these doses.

  • 2007

    First-line high-dose chemotherapy did not show benefit in trials

    Two randomised trials added high-dose chemotherapy with stem-cell rescue to first-line treatment of poor-prognosis germ-cell tumours. In one of 219 patients (intermediate or poor risk), 52% with high-dose chemotherapy and 48% with BEP alone had a durable complete response at one year, not significantly different. A 131-patient European trial, closed early, also found no significant gain.

    Motzer 2007: phase 3, 219 patients with intermediate or poor-risk germ cell tumours, BEP x4 vs BEP x2 plus high-dose carboplatin-containing chemotherapy: 1-year durable complete response 52% vs 48% (P = .53); in 67 patients with unsatisfactory marker decline 61% vs 34% (P = .03). EORTC 30974 (137 randomised, closed early; 131 analysed): complete response 44.6% vs 33.3% (P = 0.18); 2-year FFS 58.2% vs 44.8% (P = 0.060), high-dose VIP vs BEP.

Today and tomorrow

Standard of care today

Localised

Clinical stage I disease, EAU 2026 text read (Disease Management chapter, section 6.1; the page shows no update date and its download file names carry 2026). Seminoma:

  • the recommendations are to inform the person fully about surveillance and adjuvant options with their recurrence rates and short- and long-term side effects
  • to offer surveillance as the preferred option if resources are available and the person is compliant
  • to offer one dose of carboplatin AUC 7 if adjuvant chemotherapy is considered
  • not to give adjuvant treatment to people at very low risk (no risk factors)
  • and not to use adjuvant radiotherapy routinely, reserving it for highly selected people not suitable for surveillance and with a contraindication to chemotherapy. All six are rated Strong. Non-seminoma: surveillance or risk-adapted treatment based on lymphovascular invasion is recommended (Strong). For stage IA (pT1, no vascular invasion) surveillance is offered to people willing and able to comply, with one course of BEP offered to those not willing or unsuitable for surveillance
  • for stage IB (pT2-pT4) one course of BEP or surveillance is offered with the advantages and disadvantages discussed, and nerve-sparing retroperitoneal lymph node dissection is offered only to highly selected people with a contraindication to adjuvant chemotherapy who will not accept surveillance (all Strong). This entry restates the guideline and is not a ranking. The 2026 EAU summary article (Eur Urol 2026;90(2):163-174), the ESMO-EURACAN 2022 guideline and the NCCN Guidelines Insights Version 2.2025 are cited by identifier only
  • their text was not read
Guideline: European Association of Urology 2026Patrikidou A, Cazzaniga W, Oing C, Berney D, Lobo J, Boormans J, et al 2026Guideline: Oldenburg J, Berney DM, Bokemeyer C, Climent MA, Daugaard G, Gietema JA, et al 2022Guideline: Gilligan T, Lin DW, Adra N, Bagrodia A, Feldman DR, Yamoah K, et al 2025
Metastatic

First-line chemotherapy, EAU 2026 text read (sections 6.2 and 6.3.6.1). The text states that treatment depends on histology, IGCCCG prognostic group and, in poor-prognosis disease, serum tumour marker decline after the first cycle. Seminoma stage IIC and higher:

  • primary chemotherapy by IGCCCG group, BEP x 3 for good prognosis and BEP x 4 for intermediate prognosis (Strong)
  • the text describes EP x 4 as an alternative in good-risk seminoma, particularly when bleomycin is contraindicated, with similar response rates but possibly a slightly higher relapse risk, and VIP when bleomycin is contraindicated in intermediate-risk seminoma. Non-seminoma: BEP x 3 is the standard for good prognosis (EP x 4 also listed in the text's Table 6), four cycles of standard BEP for intermediate prognosis (Strong), and four cycles of BEP for poor prognosis with favourable marker decline (Strong)
  • four cycles of VIP with primary G-CSF prophylaxis is described for people with contraindications to bleomycin. Marker decline should be assessed after one cycle in poor-prognosis disease and intensification considered if it is unfavourable (Weak). Resection of visible residual masses over 1 cm after chemotherapy for non-seminoma when markers are normal or normalising is recommended (Strong). For stage IIA/B seminoma the text offers cisplatin chemotherapy by IGCCCG group or, alternatively, radiotherapy (Weak). The IGCCCG classification (1997 and the 2021 update) is cited by identifier as the system the guideline uses
  • those papers were not summarised here. This entry restates the guideline and is not a ranking
Guideline: European Association of Urology 2026International Germ Cell Cancer Collaborative Group 1997Gillessen S, Sauvé N, Collette L, Daugaard G, de Wit R, Albany C, et al 2021Beyer J, Collette L, Sauvé N, Daugaard G, Feldman DR, Tandstad T, et al 2021
Metastatic

FDA regulatory status, read through openFDA on 20 October 2026. The Drugs@FDA submission histories list original approvals (ORIG-1) of cisplatin NDA 018057 on 19 December 1978, bleomycin NDA 050443 (Blenoxane) on 31 July 1973, etoposide NDA 018768 (VePesid) on 10 November 1983 and ifosfamide NDA 019763 (Ifex) on 14 August 1987; the wording of the original indications was not read. Current labels read:

  • cisplatin for injection (WG Critical Care, NDA018057, effective 24 June 2026) is indicated for the treatment of advanced testicular cancer, with dosing of 20 mg/m2 intravenously daily for 5 days per cycle
  • etoposide injection (BluePoint, effective 20 May 2026) lists refractory testicular tumours, in combination with other approved agents, in people who have already received appropriate surgical, chemotherapeutic and radiotherapeutic therapy
  • the bleomycin label (Meitheal, effective 17 April 2026) lists testicular carcinoma (embryonal cell, choriocarcinoma, teratocarcinoma) among its indications and describes the drug as a palliative treatment. The label sets are generic products chosen from several returned by openFDA and may not match the latest DailyMed versions. The National Cancer Institute's list of drugs approved for testicular cancer (page last updated 16 May 2019) names bleomycin, cisplatin, dactinomycin, etoposide, etoposide phosphate, ifosfamide and vinblastine. No FDA or EMA action specific to germ cell tumours dated 2023-2026 was found in web searches
  • those searches were not exhaustive, no EMA document was read, and the FDA oncology approvals list was not searched directly
US Food and Drug Administration 2026US Food and Drug Administration 2026US Food and Drug Administration 2026US Food and Drug Administration 2026DailyMed 2026DailyMed 2026DailyMed 2026National Cancer Institute 2019
Relapsed / refractory
Salvage, EAU 2026 text read (section 6.3.4). The text states that cisplatin combination salvage chemotherapy results in long-term remissions in approximately 50% of people who relapse after first-line chemotherapy, with results depending strongly on prognostic factors. The regimens of choice are described as four cycles of a three-agent regimen with cisplatin and ifosfamide plus etoposide (VIP), paclitaxel (TIP) or possibly gemcitabine (GIP); the text notes that no randomised trial has compared these regimens and that they need G-CSF support and well-trained oncologists. When high-dose chemotherapy is used, the text prefers sequential high-dose carboplatin and etoposide cycles to a single high-dose regimen, reports a secondary analysis of the International Prognostic Factors Study Group cohort (n = 1,600; not a randomised comparison) showing a 10-15% overall survival improvement with high-dose over standard-dose salvage, and says that a randomised trial of high-dose versus conventional-dose chemotherapy at first relapse (TIGER) is under way. It stresses treatment within clinical trials and at specialised centres. The ClinicalTrials.gov record for TIGER (NCT02375204) shows status active, not recruiting with no results posted (checked 20 October 2026). The ifosfamide label read (Fresenius Kabi, effective 9 April 2026) states that ifosfamide is indicated in combination with certain other approved antineoplastic agents for third-line chemotherapy of germ cell testicular cancer, with mesna for prophylaxis of haemorrhagic cystitis. No germ cell tumour approval for any newer agent was found (see the FDA entry above for the limits of that search). This entry restates the guideline and is not a ranking.
Guideline: European Association of Urology 2026DailyMed 2026US Food and Drug Administration 2026ClinicalTrials 2026
Any setting
Follow-up and survivorship, EAU 2026 text read (section 7). The guideline defines three follow-up groups (stage I seminoma, stage I non-seminoma on surveillance, and people in complete remission after adjuvant treatment or chemotherapy for good- and intermediate-prognosis metastatic disease) and gives minimal schedules of tumour markers, chest X-ray and abdominopelvic MRI or CT for each in its Tables 8 to 10, based on an ESMO consensus conference; people with poor-prognosis disease or incomplete remission are to be followed individually in specialised centres. It states that MRI can replace contrast CT of the abdomen in experienced centres and that FDG-PET-CT is not recommended during surveillance. Very late relapse after five years is described as rare (approximately 0.5% in a population-based analysis), after which follow-up shifts to detecting late effects. During follow-up people should be screened and treated for hypertension, hyperlipidaemia and testosterone deficiency; adverse health outcomes are described as more common after chemotherapy than after surgery alone; and a written survivorship plan may be helpful. The ESMO-EURACAN 2022 guideline is cited by identifier only.
Guideline: European Association of Urology 2026Guideline: Oldenburg J, Berney DM, Bokemeyer C, Climent MA, Daugaard G, Gietema JA, et al 2022

Trials to watch

  • TIGER NCT02375204

    As first salvage treatment in relapsed or refractory germ cell tumors, does high-dose chemotherapy with stem cell transplant (paclitaxel and ifosfamide for mobilisation, then high-dose carboplatin and etoposide, TI-CE) differ in overall survival from conventional-dose paclitaxel, ifosfamide and cisplatin (TIP)? Registry record: randomised phase 3, primary outcome overall survival, 420 enrolled (actual). Registry status: active, not recruiting; no results posted.

  • P3BEP (ANZUP 1302) NCT02582697

    In intermediate- and poor-risk metastatic germ cell tumours, does accelerated BEP differ from standard BEP in progression-free survival (the registry's primary outcome)? Registry record: randomised phase 3, 500 estimated. Registry status: recruiting; no results posted.

  • COG AGCT1531 NCT03067181

    In children, adolescents and adults with germ cell tumors, including testicular primaries, what are overall survival and event-free survival (registry primary outcomes, two years after enrolment) with active surveillance for low-risk disease and with carboplatin versus cisplatin in a bleomycin-and-etoposide regimen for standard-risk disease? Registry record: phase 3, 1,780 estimated, not specific to testicular primaries. Registry status: recruiting; no results posted.

  • SWENOTECA-ABC NCT02341989

    In clinical stage I seminoma, how does the relapse rate at 10 years (the registry's primary outcome) compare after one course of adjuvant BEP and after one course of carboplatin AUC 7? Registry record: randomised phase 3, 348 estimated. Registry status: active, not recruiting; no results posted.

  • GIG-T NCT05874063

    In good- and intermediate-prognosis advanced germ cell tumors, does thromboprophylaxis change the proportion of people with a thromboembolic event during chemotherapy (the registry's primary outcome)? Registry record: randomised phase 3, 387 estimated. Registry status: recruiting; no results posted.

References

Show all 113 references

Loading references… Open the list as a page

Every figure was checked against its primary source to the extent possible. Items that could not be fully confirmed are marked “needs review”. Educational history, not medical advice.

Report an error

Links open the primary source. A flag marked ⚑ needs review means a value or claim has not yet been independently confirmed.

2021UnderstandingUnderstanding

majorMetastatic

IGCCCG Update: modern survival estimates and refined prognosis

Two 2021 papers re-examined the IGCCCG groups in men treated from 1990 to 2013: 9,728 with non-seminoma and 2,451 with seminoma. Five-year survival was about 96, 89 and 67 in 100 for good, intermediate and poor non-seminoma, and 95 and 88 in 100 for good and intermediate seminoma.

Pooled trial and registry data, cisplatin- and etoposide-based first-line chemotherapy, 1990-2013. NSGCT (n=9,728): 5-year OS 96% good, 89% intermediate, 67% poor; PFS 90%, 78%, 54%. Seminoma (n=2,451): 5-year OS 95% good, 88% intermediate; PFS 89%, 79%. New factors: NSGCT LDH cut-off 2.5 x ULN, age, lung metastases; seminoma LDH above 2.5 x ULN (good prognosis 3-year PFS 80% vs 92%).

Standard before

The 1997 classification, which relied on data from patients treated between 1975 and 1990.

What changed

Updated 5-year survival estimates for the same prognostic groups, a more granular prognostic model for non-seminoma that adds age and lung metastases, a new LDH cut-off of 2.5 times the upper limit of normal, and LDH above that cut-off as an adverse factor within good-prognosis seminoma.

Why it mattered

In non-seminoma the update shows higher overall survival in every group than the original estimates while progression-free survival changed little in the good and intermediate groups; in seminoma both progression-free and overall survival were higher than the original estimates. It adds prognostic factors for trial design and counselling, and an online calculator is provided.

Caveats

  • Two separate papers with different first authors: Gillessen (non-seminoma, 9,728 men) and Beyer (seminoma, 2,451 men); each has its own results record.
  • Data mix clinical-trial and registry sources from 30 institutions or collaborative groups; era (1990-2013), regimens and case mix differ from the 1975-1990 original.
  • Comparing 1997 and 2021 survival is a comparison across eras and datasets, not a treatment effect; the 1997 headline figures pool non-seminoma and seminoma, whereas the non-seminoma update compares with non-seminoma-only original values.
  • Intermediate-prognosis seminoma estimates have wide confidence intervals (for example 5-year PFS 79%, 95% CI 70 to 85).