PARADIGM
A Phase III, Randomized, Controlled Study of mFOLFOX6 + Bevacizumab Combination Therapy Versus mFOLFOX6 + Panitumumab Combination Therapy in Chemotherapy-naive Patients With KRAS/NRAS Wild-type, Incurable/Unresectable, Advanced/Recurrent Colorectal Cancer
Phase 3N = 823NCT02394795
- Population
- Chemotherapy-naive RAS wild-type unresectable metastatic colorectal cancer
- Design
- Randomised, open-label phase 3 trial at 197 sites in Japan (May 2015 to January 2022; final follow-up 14 January 2022): chemotherapy-naive patients with RAS wild-type unresectable metastatic colorectal cancer were randomised to panitumumab (n=411) or bevacizumab (n=412) plus mFOLFOX6 every 14 days. The primary endpoint, overall survival, was tested first in participants with left-sided tumours, then in the overall population. Per the full text the trial was designed to compare overall survival in all participants; the protocol was revised in May 2019, after the last participant was enrolled (June 2017) and before data were analysed, to make overall survival in left-sided tumours the primary outcome (left-sided: descending colon, sigmoid colon, rectosigmoid and rectum). 823 patients were randomised; the efficacy analyses used the as-treated population of 802 (randomised participants who received at least one dose of study treatment and met eligibility criteria; 12 received no treatment and 9 did not meet inclusion criteria), of whom 604 (75.3%) had left-sided tumours.
- Primary endpoint
- OS
- Arms
- Bevacizumab plus mFOLFOX6 (control, n = 412): Bevacizumab plus mFOLFOX6 every 14 days
- Panitumumab plus mFOLFOX6 (experimental, n = 411): Panitumumab plus mFOLFOX6 every 14 days
- People
- Watanabe J
OSHR 0.82Δ median OS+3.6 monthsOSHR 0.84 (95% CI 0.72–0.98)Δ median OS+4.9 monthsOSHR 1.09 (95% CI 0.79–1.51)Δ median OS−3 months
Show the numbers
| Endpoint | Analysis | Panitumumab plus mFOLFOX6 | Bevacizumab plus mFOLFOX6 | Effect |
|---|---|---|---|---|
| OSLeft-sided tumours (n=604 of 802), as-treated population (n=802: randomised participants who received at least one dose of study treatment and met eligibility criteria; not all randomised participants); first step of the testing sequence | primarymedian follow-up 61 monthsWatanabe 2023 | median 37.9 months | median 34.3 months | HR 0.82P=0.03Hazard ratio for death, panitumumab versus bevacizumab (stratified), with a 95.798% confidence interval of 0.68-0.99 as printed (not a 95% interval). The report states that the Kaplan-Meier curves for overall survival were not parallel and appeared to cross around 28 months, so interpretation of the observed hazard ratios is unclear. |
| OSOverall population, as-treated population (n=802: randomised participants who received at least one dose of study treatment and met eligibility criteria; not all randomised participants); second step of the testing sequence | primarymedian follow-up 61 monthsWatanabe 2023 | median 36.2 months | median 31.3 months | HR 0.84 (95% CI 0.72–0.98)P=0.03Second step of the testing sequence. The report states that the Kaplan-Meier curves for overall survival were not parallel and appeared to cross around 28 months, so interpretation of the observed hazard ratios is unclear. |
| OSRight-sided tumours (84 panitumumab, 103 bevacizumab), as-treated population (n=802: randomised participants who received at least one dose of study treatment and met eligibility criteria; not all randomised participants); predefined exploratory subgroup outside the primary testing sequence | exploratoryWatanabe 2023 | median 20.2 months (15.2–32.0) | median 23.2 months (18.5–29.1) | HR 1.09 (95% CI 0.79–1.51)Stratified hazard ratio for death, panitumumab versus bevacizumab. The report states that analyses of right-sided tumours were predefined exploratory outcomes, gives no P value for this comparison, and describes no significant overall survival difference between treatments in right-sided tumours. Participants with primary lesions on both sides (4 in the panitumumab group, 7 in the bevacizumab group) are not counted as right-sided. Median follow-up in right-sided tumours: 61.3 months (IQR 60.5-66.6) with panitumumab and 60.2 months (IQR 55.7-65.5) with bevacizumab. |
| PFSLeft-sided tumours (n=604 of 802), as-treated population (n=802: randomised participants who received at least one dose of study treatment and met eligibility criteria; not all randomised participants) | primaryWatanabe 2023 | median 13.1 months | median 11.9 months | HR 1.00 (95% CI 0.83–1.20)Secondary endpoint. |
| PFSOverall population, as-treated population (n=802: randomised participants who received at least one dose of study treatment and met eligibility criteria; not all randomised participants) | primaryWatanabe 2023 | median 12.2 months | median 11.4 months | HR 1.05 (95% CI 0.90–1.24)Secondary endpoint. |
| PFSRight-sided tumours (84 panitumumab, 103 bevacizumab), as-treated population (n=802: randomised participants who received at least one dose of study treatment and met eligibility criteria; not all randomised participants); predefined exploratory subgroup outside the primary testing sequence | exploratoryWatanabe 2023 | median 7.2 months | median 9.4 months | HR 1.43 (95% CI 1.03–1.97)Secondary endpoint, analysed here in the right-sided subgroup. The report text prints the hazard ratio without naming its direction or the word 'stratified' and gives no confidence intervals for the two medians and no P value for this comparison; a hazard ratio above 1 is consistent with the shorter median with panitumumab. The report states that findings for secondary end points should be interpreted as exploratory because no adjustment was made for multiple comparisons, and that analyses of right-sided tumours were predefined exploratory outcomes. Median follow-up for overall and progression-free survival in right-sided tumours: 61.3 months (IQR 60.5-66.6) with panitumumab and 60.2 months (IQR 55.7-65.5) with bevacizumab. |
| ORRLeft-sided tumours (n=604 of 802), as-treated population (n=802: randomised participants who received at least one dose of study treatment and met eligibility criteria; not all randomised participants) | primaryWatanabe 2023 | 80.2% | 68.6% | Secondary endpoint; difference 11.2% (95% CI 4.4%-17.9%). |
| ORROverall population, as-treated population (n=802: randomised participants who received at least one dose of study treatment and met eligibility criteria; not all randomised participants) | primaryWatanabe 2023 | 74.9% | 67.3% | Secondary endpoint; difference 7.7% (95% CI 1.5%-13.8%). |
| ORRRight-sided tumours (84 panitumumab, 103 bevacizumab), as-treated population (n=802: randomised participants who received at least one dose of study treatment and met eligibility criteria; not all randomised participants); response rates calculated on 82 and 103 participants; predefined exploratory subgroup | exploratoryWatanabe 2023 | 54.9% (43.5–65.9%) | 63.1% (53.0–72.4%) | Secondary endpoint, analysed here in the right-sided subgroup (Table 2 lists it among the exploratory outcomes in right-sided tumours): 45 of 82 with panitumumab and 65 of 103 with bevacizumab. The report prints no difference, confidence interval for a difference or P value for this comparison. |
| Endpoint | Analysis | Panitumumab plus mFOLFOX6 | Bevacizumab plus mFOLFOX6 | Effect |
|---|---|---|---|---|
| OSLeft-sided tumours (n=604 of 802), as-treated population (n=802: randomised participants who received at least one dose of study treatment and met eligibility criteria; not all randomised participants); first step of the testing sequence | primarymedian follow-up 61 monthsWatanabe 2023 | median 37.9 months | median 34.3 months | HR 0.82P=0.03Hazard ratio for death, panitumumab versus bevacizumab (stratified), with a 95.798% confidence interval of 0.68-0.99 as printed (not a 95% interval). The report states that the Kaplan-Meier curves for overall survival were not parallel and appeared to cross around 28 months, so interpretation of the observed hazard ratios is unclear. |
| OSOverall population, as-treated population (n=802: randomised participants who received at least one dose of study treatment and met eligibility criteria; not all randomised participants); second step of the testing sequence | primarymedian follow-up 61 monthsWatanabe 2023 | median 36.2 months | median 31.3 months | HR 0.84 (95% CI 0.72–0.98)P=0.03Second step of the testing sequence. The report states that the Kaplan-Meier curves for overall survival were not parallel and appeared to cross around 28 months, so interpretation of the observed hazard ratios is unclear. |
| OSRight-sided tumours (84 panitumumab, 103 bevacizumab), as-treated population (n=802: randomised participants who received at least one dose of study treatment and met eligibility criteria; not all randomised participants); predefined exploratory subgroup outside the primary testing sequence | exploratoryWatanabe 2023 | median 20.2 months (15.2–32.0) | median 23.2 months (18.5–29.1) | HR 1.09 (95% CI 0.79–1.51)Stratified hazard ratio for death, panitumumab versus bevacizumab. The report states that analyses of right-sided tumours were predefined exploratory outcomes, gives no P value for this comparison, and describes no significant overall survival difference between treatments in right-sided tumours. Participants with primary lesions on both sides (4 in the panitumumab group, 7 in the bevacizumab group) are not counted as right-sided. Median follow-up in right-sided tumours: 61.3 months (IQR 60.5-66.6) with panitumumab and 60.2 months (IQR 55.7-65.5) with bevacizumab. |
| PFSLeft-sided tumours (n=604 of 802), as-treated population (n=802: randomised participants who received at least one dose of study treatment and met eligibility criteria; not all randomised participants) | primaryWatanabe 2023 | median 13.1 months | median 11.9 months | HR 1.00 (95% CI 0.83–1.20)Secondary endpoint. |
| PFSOverall population, as-treated population (n=802: randomised participants who received at least one dose of study treatment and met eligibility criteria; not all randomised participants) | primaryWatanabe 2023 | median 12.2 months | median 11.4 months | HR 1.05 (95% CI 0.90–1.24)Secondary endpoint. |
| PFSRight-sided tumours (84 panitumumab, 103 bevacizumab), as-treated population (n=802: randomised participants who received at least one dose of study treatment and met eligibility criteria; not all randomised participants); predefined exploratory subgroup outside the primary testing sequence | exploratoryWatanabe 2023 | median 7.2 months | median 9.4 months | HR 1.43 (95% CI 1.03–1.97)Secondary endpoint, analysed here in the right-sided subgroup. The report text prints the hazard ratio without naming its direction or the word 'stratified' and gives no confidence intervals for the two medians and no P value for this comparison; a hazard ratio above 1 is consistent with the shorter median with panitumumab. The report states that findings for secondary end points should be interpreted as exploratory because no adjustment was made for multiple comparisons, and that analyses of right-sided tumours were predefined exploratory outcomes. Median follow-up for overall and progression-free survival in right-sided tumours: 61.3 months (IQR 60.5-66.6) with panitumumab and 60.2 months (IQR 55.7-65.5) with bevacizumab. |
| ORRLeft-sided tumours (n=604 of 802), as-treated population (n=802: randomised participants who received at least one dose of study treatment and met eligibility criteria; not all randomised participants) | primaryWatanabe 2023 | 80.2% | 68.6% | Secondary endpoint; difference 11.2% (95% CI 4.4%-17.9%). |
| ORROverall population, as-treated population (n=802: randomised participants who received at least one dose of study treatment and met eligibility criteria; not all randomised participants) | primaryWatanabe 2023 | 74.9% | 67.3% | Secondary endpoint; difference 7.7% (95% CI 1.5%-13.8%). |
| ORRRight-sided tumours (84 panitumumab, 103 bevacizumab), as-treated population (n=802: randomised participants who received at least one dose of study treatment and met eligibility criteria; not all randomised participants); response rates calculated on 82 and 103 participants; predefined exploratory subgroup | exploratoryWatanabe 2023 | 54.9% (43.5–65.9%) | 63.1% (53.0–72.4%) | Secondary endpoint, analysed here in the right-sided subgroup (Table 2 lists it among the exploratory outcomes in right-sided tumours): 45 of 82 with panitumumab and 65 of 103 with bevacizumab. The report prints no difference, confidence interval for a difference or P value for this comparison. |
Data table
| Endpoint | HR | 95% CI |
|---|---|---|
| OS | 0.84 | 0.72–0.98 |
| OS | 1.09 | 0.79–1.51 |
| PFS | 1 | 0.83–1.2 |
| PFS | 1.05 | 0.9–1.24 |
| PFS | 1.43 | 1.03–1.97 |
- Panitumumab plus mFOLFOX6 (experimental)
- Bevacizumab plus mFOLFOX6 (control)
Median-only curves assume a constant hazard (exponential) and fit plateaus, as seen with some immunotherapies, poorly.
Data table (modelled values)
| Months | Panitumumab plus mFOLFOX6 | Bevacizumab plus mFOLFOX6 |
|---|---|---|
| 0 | 100.0% | 100.0% |
| 24 | 64.5% | 61.6% |
| 47 | 42.3% | 38.7% |
| 71 | 27.3% | 23.8% |
| 95 | 17.6% | 14.7% |
- Panitumumab plus mFOLFOX6 (experimental)
- Bevacizumab plus mFOLFOX6 (control)
Median-only curves assume a constant hazard (exponential) and fit plateaus, as seen with some immunotherapies, poorly.
Data table (modelled values)
| Months | Panitumumab plus mFOLFOX6 | Bevacizumab plus mFOLFOX6 |
|---|---|---|
| 0 | 100.0% | 100.0% |
| 23 | 64.4% | 60.1% |
| 45 | 42.2% | 36.9% |
| 68 | 27.2% | 22.2% |
| 91 | 17.5% | 13.3% |
Appears in
Sources
Watanabe J, Muro K, Shitara K, Yamazaki K, Shiozawa M, Ohori H, et al. Panitumumab vs Bevacizumab Added to Standard First-line Chemotherapy and Overall Survival Among Patients With RAS Wild-type, Left-Sided Metastatic Colorectal Cancer: A Randomized Clinical Trial. JAMA. 2023;329(15):1271-1282.