Long-term effect of a watch and wait policy versus immediate systemic treatment for asymptomatic advanced-stage non-Hodgkin lymphoma: a randomised controlled trial
N = 309
- Population
- Asymptomatic, advanced-stage, low-grade non-Hodgkin lymphoma (the abstract does not restrict the population to follicular histology); 158 randomised to immediate chlorambucil and 151 to observation.
- Design
- Randomised controlled trial in 44 UK centres (recruitment 1 February 1981 to 31 July 1990) in asymptomatic patients with advanced-stage, low-grade non-Hodgkin lymphoma: immediate oral chlorambucil 10 mg per day continuously versus an initial policy of observation with systemic therapy delayed until disease progression. Local radiotherapy to symptomatic nodes was allowed in both groups. 309 patients.
- Arms
- Immediate chlorambucil (control, n = 158): Oral chlorambucil 10 mg per day continuously, started at randomisation
- Observation (watch and wait) (experimental, n = 151): Initial policy of observation, with systemic therapy delayed until disease progression
- People
- Ardeshna KM
Δ median OS+0.8 yearsΔ median Cause-specific survival+0.1 years
Show the numbers
| Endpoint | Analysis | Observation (watch and wait) | Immediate chlorambucil | Effect |
|---|---|---|---|---|
| OSRandomised patients; analysis population not labelled in the abstract | long term-updatemedian follow-up 16 yearsArdeshna 2003 | median 6.7 years | median 5.9 years | P=0.84Ranges as published: 0 to 17.8 years with chlorambucil and 0.5 to 18.9 years with observation. The abstract reports no hazard ratio. |
| Cause-specific survivalRandomised patients; analysis population not labelled in the abstract | long term-updatemedian follow-up 16 yearsArdeshna 2003 | median 9.1 years | median 9 years | P=0.44Ranges as published: 0 to 17.8 years and 0.67 to 18.9 years. |
| Time to new treatmentObservation group | long term-updateArdeshna 2003 | 19% at 10 y | — | Actuarial chance of not needing chemotherapy at 10 years of follow-up, with non-lymphoma deaths censored; the abstract gives 40% for patients older than 70 years. Reported for the observation group only. |
| Endpoint | Analysis | Observation (watch and wait) | Immediate chlorambucil | Effect |
|---|---|---|---|---|
| OSRandomised patients; analysis population not labelled in the abstract | long term-updatemedian follow-up 16 yearsArdeshna 2003 | median 6.7 years | median 5.9 years | P=0.84Ranges as published: 0 to 17.8 years with chlorambucil and 0.5 to 18.9 years with observation. The abstract reports no hazard ratio. |
| Cause-specific survivalRandomised patients; analysis population not labelled in the abstract | long term-updatemedian follow-up 16 yearsArdeshna 2003 | median 9.1 years | median 9 years | P=0.44Ranges as published: 0 to 17.8 years and 0.67 to 18.9 years. |
| Time to new treatmentObservation group | long term-updateArdeshna 2003 | 19% at 10 y | — | Actuarial chance of not needing chemotherapy at 10 years of follow-up, with non-lymphoma deaths censored; the abstract gives 40% for patients older than 70 years. Reported for the observation group only. |
- Immediate chlorambucil (control)
- Observation (watch and wait) (experimental)
Median-only curves assume a constant hazard (exponential) and fit plateaus, as seen with some immunotherapies, poorly.
Data table (modelled values)
| Months | Immediate chlorambucil | Observation (watch and wait) |
|---|---|---|
| 0 | 100.0% | 100.0% |
| 50 | 61.3% | 65.0% |
| 101 | 37.2% | 41.9% |
| 151 | 22.8% | 27.2% |
| 201 | 14.0% | 17.7% |
- Immediate chlorambucil (control)
- Observation (watch and wait) (experimental)
Median-only curves assume a constant hazard (exponential) and fit plateaus, as seen with some immunotherapies, poorly.
Data table (modelled values)
| Months | Immediate chlorambucil | Observation (watch and wait) |
|---|---|---|
| 0 | 100.0% | 100.0% |
| 68 | 64.6% | 64.9% |
| 137 | 41.5% | 41.9% |
| 205 | 26.8% | 27.2% |
| 273 | 17.3% | 17.7% |