Adjuvant Chemoradiotherapy versus Radiotherapy Alone for Women with High-Risk Endometrial Cancer (PORTEC-3)

Final Results of an International, Open-Label, Multicentre, Randomised, Phase 3 Trial

de Boer SM, Powell ME, Mileshkin L, et al.
Lancet Oncology 2018; 19: 295-309

Background & Rationale

 

  • Women with high-risk endometrial cancer (15% of cases) have increased risk of recurrence and cancer-related death
  • Standard adjuvant treatment has been pelvic external beam radiotherapy (EBRT)
  • Radiotherapy shown to delay pelvic recurrence but not improve overall survival
  • Chemotherapy shown to delay distant metastases but also without survival benefit
  • RTOG 9708 phase 2 trial explored combination of EBRT with concurrent and adjuvant chemotherapy with promising results
  • PORTEC-3 trial initiated to determine whether addition of chemotherapy to radiotherapy improves outcomes in high-risk endometrial cancer

Study Design

686 patients with high-risk endometrial cancer randomized 1:1
↓
Radiotherapy Alone
↓
EBRT (48.6 Gy in 1.8 Gy fractions)
5 days/week
Chemoradiotherapy
↓
EBRT (48.6 Gy) + concurrent cisplatin
50 mg/m² in week 1 & 4
↓
Adjuvant chemotherapy
4 cycles carboplatin AUC5 + paclitaxel 175 mg/m²

 

  • Open-label, international, randomized phase 3 trial
  • 103 centers in 6 clinical trial groups collaborating in the Gynaecological Cancer Intergroup
  • Randomization stratified by participating center, lymphadenectomy, stage, and histological type
  • Co-primary endpoints: overall survival and failure-free survival
  • Vaginal brachytherapy boost allowed if cervical involvement (46% in chemoRT vs 48% in RT)

Eligibility Criteria

  • Endometrioid endometrial cancer:
FIGO 2009 stage IA grade 3 with documented LVSI
FIGO stage IB grade 3
FIGO stage II
FIGO stage IIIA, IIIB (parametrial), or IIIC
  • Serous or clear cell histology:
Stage IA with invasion
Stage IB, II, or III
  • Additional criteria:
WHO performance status 0-2
Adequate bone marrow, liver, and kidney function
Age ≥18 years (no upper limit)

Patient Characteristics

  • Median age: 62 years (56-68)
  • FIGO 2009 stage:
    • Stage I: 30%
    • Stage II: 25%
    • Stage III: 45%
  • Histological type:
    • Endometrioid grade 1: 19%
    • Endometrioid grade 2: 20%
    • Endometrioid grade 3: 28%
    • Serous: 16%
    • Clear cell: 9%
    • Mixed/other: 8%
  • Myometrial invasion ≥50%: 64%
  • LVSI present: 59%
  • WHO performance score 0-1: 99%
  • Surgery:
    • Total abdominal/laparoscopic hysterectomy with BSO: 29%
    • Total abdominal/laparoscopic hysterectomy with BSO + lymphadenectomy/staging: 71%
  • Median nodes removed when lymphadenectomy performed: 14-15

660 eligible patients analyzed (330 in each arm)
Baseline characteristics well balanced between groups

Patient Characteristics

Treatment Completion

Treatment Chemoradiotherapy
n=330		 Radiotherapy
n=330
EBRT completed 329 (100%) 325 (99%)
Vaginal brachytherapy boost 151 (46%) 158 (48%)
2 cycles cisplatin completed 304 (92%) NA
4 cycles carboplatin completed 262 (79%) NA
4 cycles paclitaxel completed 233 (71%) NA
  • Chemotherapy was discontinued in 61 patients (18%):
    • Due to toxicity: 31 (9%)
    • Patient decision: 20 (6%)
    • Disease progression: 7 (2%)
    • Other reasons: 3 (1%)
  • Dose reductions:
    • Cisplatin (40 mg/m²): 5 patients (2%)
    • Carboplatin (AUC5 to AUC4): 36 patients (11%)
    • Paclitaxel (175 mg/m² to 135 mg/m²): 50 patients (15%)

Survival Outcomes

Outcome at 5 years Chemoradiotherapy
n=330		 Radiotherapy
n=330		 Hazard Ratio (95% CI) p value*
Overall survival 81.8% (77.5-86.2) 76.7% (72.1-81.6) 0.76 (0.54-1.06) 0.109
Failure-free survival 75.5% (70.3-79.9) 68.6% (63.1-73.4) 0.71 (0.53-0.95) 0.022

*Adjusted for stratification factors

Survival Outcomes

Outcome at 5 years Chemoradiotherapy
n=330		 Radiotherapy
n=330		 Hazard Ratio (95% CI) p value*
Overall survival 81.8% (77.5-86.2) 76.7% (72.1-81.6) 0.76 (0.54-1.06) 0.109
Failure-free survival 75.5% (70.3-79.9) 68.6% (63.1-73.4) 0.71 (0.53-0.95) 0.022

*Adjusted for stratification factors

Recurrence Patterns

Recurrence Type Chemoradiotherapy
n=330		 Radiotherapy
n=330		 p value
First recurrence
Vaginal recurrence 0.3% 0.3% 0.999
Pelvic recurrence 1.0% 1.5% 0.473
Distant metastases 22.4% 28.3% 0.108
Total recurrence (includes multiple sites)
Vaginal recurrence 2.1% 2.1% 0.995
Pelvic recurrence 4.9% 9.2% 0.026
Distant metastases 23.1% 29.7% 0.077
  • Most recurrences were distant metastases in both arms
  • Isolated vaginal and pelvic recurrences were rare
  • Significant reduction in total pelvic recurrences with chemoradiotherapy
  • Trend toward reduction in distant metastases with chemoradiotherapy

Subgroup Analysis - Stage III

Outcome by Stage Chemoradiotherapy Radiotherapy Hazard Ratio (95% CI) p value
Stage III (n=295)
5-year overall survival 78.7% (72.2-85.7) 69.8% (62.4-78.1) 0.71 (0.45-1.11) 0.074*
5-year failure-free survival 69.3% (61.1-76.2) 58.0% (49.3-65.7) 0.66 (0.45-0.97) 0.014*
Stage I-II (n=365)
5-year failure-free survival 80.8% (74.1-86.0) 76.6% (69.5-82.2) 0.85 (0.54-1.33) 0.47

*Adjusted for stratification factors

Subgroup Analysis - Serious

Outcome by Histology Chemoradiotherapy Radiotherapy Hazard Ratio (95% CI) p value
Serous Cancer (n=105)
5-year overall survival 71.4% (60.1–84.7) 52.8% (40.6–68.6) 0.48 (0.24–0.96) 0.037
5-year failure-free survival 59.7% (45.1–71.6) 47.9% (33.9–60.6) 0.42 (0.22–0.80) 0.008
Recurrence risk: 44.8% in serous cancers vs 27.7% in grade 3 endometrioid and 27.4% in clear cell cancers

Subgroup Analysis - Mutation

Prognostic Factors - Multivariable Analysis

Only age group was found to be predictive of treatment effect (pinteraction=0.012)
Women aged ≥70 years had the greatest benefit from chemoradiotherapy

Prognostic Factors - Multivariable Analysis

Prognostic Factor 5-year OS
HR (95% CI)		 p value 5-year FFS
HR (95% CI)		 p value
Age ≥70 vs <60 years 3.29 (1.99-5.44) <0.0001 2.14 (1.41-3.25) <0.0001
Age 60-69 vs <60 years 2.31 (1.48-3.59) <0.0001 1.74 (1.23-2.46) <0.0001
Stage III vs I-II 2.41 (1.66-3.51) <0.0001 2.62 (1.90-3.61) <0.0001
Grade 3 endometrioid vs grade 1-2 1.76 (1.10-2.81) <0.0001 1.56 (1.06-2.30) <0.0001
Serous/clear cell vs grade 1-2 endometrioid 2.35 (1.48-3.72) <0.0001 2.15 (1.46-3.16) <0.0001
LVSI present vs absent 1.36 (0.93-1.98) 0.11 1.36 (0.99-1.87) 0.054
Only age group was found to be predictive of treatment effect (pinteraction=0.012)
Women aged ≥70 years had the greatest benefit from chemoradiotherapy

Prognostic Factors 

Only age group was found to be predictive of treatment effect (pinteraction=0.012)
Women aged ≥70 years had the greatest benefit from chemoradiotherapy

Toxicity During Treatment

Adverse Events Chemoradiotherapy
n=330		 Radiotherapy
n=330		 p value
Grade 2 or worse 308 (93%) 144 (43%) <0.0001
Grade 3 or worse 198 (60%) 41 (12%) <0.0001
Selected Grade 3-4 Adverse Events
Hematological 149 (45%) 18 (5%) <0.0001
Gastrointestinal 47 (14%) 18 (5%) <0.0001
Neuropathy 23 (7%) 0 (0%) <0.0001
Pain 31 (9%) 4 (1%) <0.0001
  • No treatment-related deaths
  • Most adverse events were hematological
  • Neuropathy grade 2+ was reported in 25% of patients in chemoradiotherapy group
  • From 12 months onwards, no significant differences in grade 3-4 adverse events
  • However, grade 2+ sensory neuropathy persisted in chemoradiotherapy group:
    • 8% vs 1% at 3 years (p<0.0001)
    • 9% vs 0% at 5 years (p<0.0001)

Toxicity After Treatment

Adverse Events 6 Months Post-Treatment 3 Years 5 Years
  ChemoRT RT ChemoRT RT ChemoRT RT
Grade 2 or worse events
Any 39% 29% 32% 24% 40% 28%
Sensory neuropathy 12% 0% 8% 1% 9% 0%
Gastrointestinal 6% 5% Detailed data for individual toxicities at 3 and 5 years not reported in this paper
Hematological 16% 8%
Fatigue 3% 1%
Pain 9% 10%
Alopecia 19% 0%
Hypertension 5% 5%
Grade 3 or worse events
Any 16% 8% No significant differences between groups from 12 months onwards

Data from de Boer et al. Lancet Oncology 2018

Conclusions

  • Addition of chemotherapy to radiotherapy improved 5-year failure-free survival (75.5% vs 68.6%, p=0.022) but not overall survival (81.8% vs 76.7%, p=0.109)
  • Greatest benefit seen in patients with:
    • Stage III disease
    • Age ≥70 years
  • No significant benefit observed for stage I-II disease
  • Significantly higher rates of toxicity during treatment with chemoradiotherapy:
    • Grade 3-4 adverse events: 60% vs 12%
    • Grade 2+ sensory neuropathy persisted long-term
  • Pelvic control was high with radiotherapy alone (vaginal recurrence ~2%, pelvic recurrence ~9%)
  • Majority of recurrences were distant metastases in both arms

Strengths and Limitations

Strengths

  • Large, international, phase 3 trial
  • Strong collaborative effort across 6 clinical trial groups
  • Balanced baseline characteristics
  • Clear eligibility criteria
  • High treatment compliance
  • Central pathology review
  • Comprehensive toxicity assessment
  • Inclusion of elderly patients

Limitations

  • Open-label design (not blinded)
  • Lower event rates than expected
  • Final analysis time-based rather than event-based
  • Variable use of lymphadenectomy (58% of patients)
  • Heterogeneous population (different histologies and stages)
  • Insufficient power to detect small differences in overall survival
  • No quality of life data presented in this publication

Clinical Implications

  • Combined chemoradiotherapy cannot be recommended as new standard for all high-risk endometrial cancer patients
  • For stage III disease, chemoradiotherapy should be considered to maximize failure-free survival
  • For stage I-II disease, radiotherapy alone provides excellent pelvic control without the added toxicity of chemotherapy
  • Benefits and risks of combined treatment should be individually discussed with each patient
  • Age should not be a limiting factor - elderly patients derived substantial benefit
  • Long-term risk of persistent neuropathy should be balanced against potential benefits
  • Further follow-up needed to evaluate long-term survival outcomes

Discussion Points

  1. How do the PORTEC-3 results compare with other trials in high-risk endometrial cancer (GOG-258, GOG-249, NSGO-EC-9501/EORTC-55991)?
  2. Would a different chemotherapy regimen or sequence (e.g., sequential rather than concurrent) have resulted in different outcomes?
  3. How should we approach treatment decisions for specific histological subtypes, particularly serous cancers?
  4. What is the role of molecular classification (POLE, p53, MSI) in treatment decision-making for high-risk endometrial cancer?
  5. How do we balance the improved failure-free survival in stage III disease against the increased toxicity of combined treatment?

Subsequent Publications from PORTEC-3

  • de Boer et al. (Lancet Oncol 2019): Toxicity and quality-of-life results
    • Significantly higher symptom burden during and 3 months after treatment
    • Rapid recovery of HRQOL after treatment completion
    • Persistent sensory neuropathy and tingling in extremities
  • León-Castillo et al. (JCO 2020): Molecular risk classification
    • Benefit of chemoradiotherapy limited to p53-abnormal cancers
    • No significant benefit in NSMP, MMRd, or POLE-mutant tumors
    • POLE-mutant tumors had excellent outcomes in both arms
  • Wortman et al. (IJROBP 2021): Comparison of IMRT vs. 3DCRT
    • Significantly less grade 2+ GI toxicity with IMRT (15% vs. 4%)
    • Improved patient-reported bowel symptoms with IMRT