Long Term Evolution of 328 Stage A0 B-CLL

Because at present time stages A0, i.e. stages A without peripheral tumoral syndrome, represent more than 50% of new diagnosed B-CLLs, we took back our 328 patients (pts) seen before 1998.

Sex ratio-M/F is 1.41 similar to other pts. If a roughly “systematic” numeration leads to diagnosis of 90% of pts, 6% of our pts present B-symptoms. Median time between diagnosis and first pathologic numeration is 17 months and median value of LDT is 36 months with only 11% of pts with a LDT less than 12 months. Mean values of blood and medullary lymphocytosis are 19.5 G per l and 50% and among 222 BM biopsies, a diffuse dense pattern is seen in 8% of cases. Mean levels of hemoglobin and platelets are 141 g (5 pts with stage 3 Rai) and 243 G. Mean levels of seric LDH and β2-m are 0.96 N and 2.23 mg. Stage A’’ according to the French definition represent only 19% of our pts while around 40% of our pts present at less one of our 5 clinical pejorative factors: 7 have a blood lymphocytosis >70 G; 70 a massive marrow involvement (>70% at myelogramm or diffuse pattern at BM); 25 a hemoglobin <120 g and 25 platelets <150 G, the 5^th factor being the presence of a peripheral tumoral syndrome.

At time of our first analysis (2006 July), the median follow-up time was more than 11 years and the median survival time was 9.7 years while 15- and 20-year survival rates were 33 and 20%. Median “corrected” survival time (without not-related deaths including those due to a solid tumor) was 15.4 years and the 20-year survival rate was 47% not statistically different from a matched control population. Only 3 pts developed a secondary MDS/AML while 15-year rates for occurrence of a Richter’s syndrome and a solid tumor are 10% and 25%.

If about 1/3 of our pts were treated at diagnosis, the 239 other ones were watched and only 20% of them needed a treatment. Median time of treatment is 65 months and the 5-and 10-year rates are 17 and 31% without aspect of plateau on actuarial curve. The 5-year rates of treatment are 17 and 31% among A′ and A″ pts (p=0.01) but our prognostic classification is more efficient with 13, 38 and 100% rates (p<10–6). Presence of LDH>1.25 N is also an indicator for treatment: 5-year rates are respectively 35 and 18% (p=0.04) and presence of β2-m >3 mg too: 35 and 18% (p=0.002). Combination of LDH and β2-m is more efficient: 5-year rates are 43 and 16% (p=0.0005) and this prognostic system is independent from our prognostic classification.

Pts initially treated present a smaller “corrected” survival than watched pts: median are respectively 9 and 10.3 years and 15-year rates are 26 and 37% (p=0,006). Non related deaths represent more than half of all deaths and among them solid tumor represent 41% of causes before cardio-vascular causes and age (35 and 22%).

We will actualize this study but 2 points are already clear. The “old” clinical prognostic factors remain very pertinent among these “good” B-CLL and need to be compared with “modern” prognostic factors. At least looking at treatment, primum non nocere.