Page 169 - Haematologica - Vol. 105 n. 6 - June 2020

IGHV and outcome in CLL: a population-based study
233 received second-line treatment during follow-up. Among patients who received second-line treatment, the mutational status was known for 167 (72%), of whom 117 (70%) had U-CLL. Of these unmutated cases, 33 (28%) received targeted treatment, compared with six (12%) of mutated patients.
Discussion
We present real-world data on the prognosis of CLL, from diagnosis and from time of first-line treatment, based on IGHV mutational status from the hitherto largest nationwide, population-based cohort. The main novel finding is the lengthy OSt of both M-CLL and U- CLL patients treated with either FCR or BR, despite infe- rior TFSt of U-CLL patients; reflecting U-CLL patients’ response to salvage treatment. We confirm previous find- ings of U-CLL being associated with shorter OSd and TFSd compared with M-CLL.12-14
In clinical trial reporting, progression-free survival (PFS), defined as the time until death or disease progression, is commonly used to evaluate treatment outcome. Previous studies have reported superior PFS for M-CLL patients compared with U-CLL,5,6,16,17 and M-CLL patients without del(17p) or del(11q) in particular have long PFS upon chemoimmunotherapy.17 The 3-year PFS rates reported in previous clinical trials with FCR (66-83%)5,6,16,22-24 or BR (77% for M-CLL)6,25 and in the real-world setting with FCR (60-80%),17,26 are high, but inferior to those achieved with ibrutinib-based treatment (83-96%).22,25,27,28 Upon chloram- bucil treatment the median PFS was 9 months in a Swedish real-world setting26 and 11 months in clinical trials,29,30 com- pared with 1.3-2.4 years upon CD20-chlorambucil treat- ment in clinical trials.29-32
Here we report the data for treatment-free survival, which can be clinically more relevant as many patients do not meet the International Workshop on CLL criteria for treatment at disease progression.33 Our results reveal a superior TFSt for M-CLL patients compared with U-CLL patients, when treated with FCR, BR, or non-intensive treatment regimens. We observed high 3-year TFSt rates for M-CLL patients treated with FCR (90%) or BR (91%) and for U-CLL patients treated with FCR (76%), similar to the findings of a smaller retrospective study of BR- treated patients who had a 3-year TFSt of 90%.34 This is especially impressive considering the median age of 70 years of the patients treated with BR in our study. Only 6% of M-CLL patients in our study had high-risk cytoge- netics, consistent with the long TFSt of M-CLL patients that we observed.
A 3-year OSt of 86-88% was demonstrated for patients treated with either FCR or BR. These findings are compa- rable with those from randomized clinical trials of FCR (84-91%)5,6,16,22-24 and BR (89-92%),6,25,35 and other real- world studies of FCR-treated patients (83-95%).17,26 However, the overall superior OSt for M-CLL patients previously reported as a dichotomous variable4-6,16,17 and as a continuous variable,36 was not observed in our study. This could reflect that the follow-up time in our study may have been too short for differences in OSt to mani- fest, and that factors such as comorbidity and subsequent lines of treatment may have been unevenly distributed across groups. Furthermore, a tendency to favor FCR over other treatment options for patients with U-CLL was
observed. This might reflect physicians’ choice of treat- ment intensification based on recognition of the inferior prognosis for this group of patients. More than twice as many U-CLL patients received second-line treatment and more than five times as many were given targeted agents compared with M-CLL patients. This is likely due to the shorter TFSt of U-CLL patients, again in recognition of their inferior prognosis, and may in part explain why no difference in OSt was observed.
A previous study demonstrated that the prognostic impact of IGHV status upon chemoimmunotherapy is not driven by a difference in complete remission rate, but rather by earlier relapse due to the aggressiveness of the disease in U-CLL patients.18 The long OSt after treatment with FCR or BR for both U-CLL and M-CLL patients in our study emphasizes that patients who progress after first-line chemoimmunotherapy may be salvaged with targeted treatment, or even repeated chemoimmunother- apy. A recent conference presentation described superior PFS, improved OS, and less toxicity with ibrutinib plus rituximab compared with FCR; however, subgroup analyses indicated that the benefit was mainly for U-CLL patients.22 The findings were similar, also mainly with impact on U-CLL patients, for ibrutinib-based regimens in comparison with BR, although without a difference in OS.25 Cross-over was not allowed in these studies; thus, it remains to be systematically assessed in a clinical trial whether patients with progressive disease may be sal- vaged with targeted agents in second-line treatment. In view of the long period off treatment in general, and the possibility of a clinical cure for a substantial subgroup of M-CLL patients, chemoimmunotherapy remains a legiti- mate treatment option with robust data on safety and long-term outcome in the era of targeted agents. Thus, we suggest that intensive chemoimmunotherapy should be part of a personalized treatment landscape in CLL alongside targeted treatment, with treatment options adapted based on shared decision-making, guided by robust data from clinical trials and real-world evidence.
We found that patients treated with chlorambucil as monotherapy had a poor TFSt regardless of IGHV status. Within 1 year of initiation of first-line treatment with chlorambucil, 50% of the patients had had an event. Patients treated with CD20-chlorambucil, with a median age of 78 years and representing a frail patient population, had a median TFSt of 2.5 years. A similar median TFSt of 3.4 years was observed in patients of comparable age receiving CD20-chlorambucil in randomized controlled trials.37 The superior outcome of patients treated with CD20-chlorambucil, compared with chlorambucil sug- gests that chlorambucil as monotherapy must be consid- ered obsolete. As baseline characteristics were similar between the groups in our study, our findings indicate that chlorambucil should be replaced by CD20-chloram- bucil or by more effective and less toxic novel treatments. Considering the results of our study, as well as those of clinical trials, new treatment options and improved sup- portive care are warranted for patients unfit for intensive chemoimmunotherapy. The recently published studies on venetoclax plus obinutuzumab and ibrutinib-based frontline therapy, compared with CD20-chlorambucil, broaden the options for these patients. Subgroup findings indicate that mainly U-CLL patients benefit from targeted therapy, as also seen when compared with intensive chemoimmunotherapy regimens.25,32,38,39
haematologica | 2020; 105(6)
1627