Longevity studies-PV
I had promised to look at a couple of the larger longevity studies in relation to PV, so, briefly, here they are.
"POLYCYTHEMIA VERA: THE NATURAL HISTORY OF 1213 PATIENTS FOLLOWED FOR 20 YEARS" by Gruppo Italiano Studio Policitemia publ. in Annals of Internal Medicine, 1995; vol. 123. This Italian retrospective study was conducted by sending questionnaires to doctors at 11 different institutions. There are some obvious drawbacks to conclusions that can be drawn from this type of study, and I'll mention those below. First a note on the data that was collected. The median age of diagnosis for men was 60; for women was 62. The mean follow-up time was 5.3 years. There was a break-down of patient age; 82 (7%) were said to be younger than 40 at diagnosis, 490 (40%) were 40 to 59 years, 400 (33%) were 60 to 70 years, and 241 (20%) were over 70. In survival analysis - patients over 70 had a median survival of 8 years; 69% of patients aged 40 to 60 years at diagnosis and more than 90% of those younger than 40 years survived more than 15 years. During the study there were 225 deaths (18.5%) - these are reported as thrombotic complications in 30%; acute leukemia in 15%; other cancers - 15%. PV complications are said to be low. Now for the difficulties with the study - although an attempt is made to identify thrombotic events by age and by treatment - spent phase or MF is not. And, as with the PVSG studies, the phlebotomy-only group pretty much disappears after a few years. In other words, though 373 started with no treatment, by the 3rd year it was down to 161 and so on until at the 15 years there were only 9 still on phlebotomy-only. (The study actually quotes the PVSG results for survival based on treatment.) A letter to the editor in a subsequent issue of the journal, written by two doctors from the University of Pittsburgh, offers three particular criticisms of why they believe the retrospective design of the study is faulty and presents a "somewhat" (my word) longer survival than non-retrospective studies. By questioning doctors "after the fact" - 1) sudden deaths before diagnosis or undiagnosed deaths are not included; 2) doctors frequently will point to an earlier date for disease onset than when the disease is actually diagnosed - also giving longer life-spans than clinical trial studies 3) the Italian doctors concluded that myelosuppression had an overall unfavourable effect and the PA doctors pointed out that since disease was frequently severe (or had become severe) in the myelosuppressed patients - it was a "biased assessment." The Italian doctors answer these criticisms - basically by agreeing that these problems exist with the restrospective design and emphasize that their main purpose in the trial was to provide information to set up clinical trials with low-dose aspirin (as per thrombotic events in PV). (This has been done, by the way. Berlin and Berk, in the final PVSG report, lament that the future of PV research seems to lie away from North America, in Europe.) I believe another problem not mentioned by the PA docs (maybe it's not "scientific" enough) is in the very nature of what the Italian group was looking for - thrombotic events (not spent phase) - and in doctors ability to recognize or note these changes in patients. (We've had a number come online without a clue of what had happened to them in respect to spent phase PV. In some cases, perhaps their doctors just didn't tell them; but in others, it's obvious that the docs didn't have a clue either. And, in another Italian study below, the criteria for spent phase are rather advanced, more like full-blown MF than spent.) And, one could also look at the old PVSG criteria for PV which is what is being used here (and, again, dependent on these various docs in having followed it). Najean and others (Messinezy & Pearson, for example) state that the LAP and B12 are of doubtful value in PV diagnosis. So....it's a fine study in respect to thrombotic events, which is what the authors intended. Survival (though higher as per the retrospective design) is consistent with what we have seen and heard elsewhere, and younger are doing better than older in total number of years.
LIFE EXPECTANCY OF PATIENTS WITH CHRONIC NONLEUKEMIC MYELOPROLIFERATIVE DISORDERS by Rozman, et al., publ. in Cancer, 1991. This Spanish study included 454 PV patients from 38 institutions. It was apparently retrospective so the same criticisms of design, as per the previous study, could be applied. (I guess.) The median age was 62 and the mean follow-up was 52 months. The study lists survival charts for the 3 diseases (PV, ET and MF). The PV charts do not significantly differ from the general expected survival of the population for the first 10 years; not quite as good after that. (Though these Spanish survival expectations do not seem quite as high as recent ones listed for North America, I believe. Survival expectations do go up after one reaches certain ages - for example, higher at 60/65 or whatever than if just calculated from 45. Whether this is reflected in the study by the fact that after 10 years of PV, the stats don't fare as well with the general population, I don't know. I'm not a statistician!) The obvious problem with the study in respect to PV? Well, fine if you're at the median age of 62 - doesn't give us much to go on for younger PVers. AND, again, this study goes to the PVSG stats when it comes to listing survival as per treatment.
So how about studies with younger PV patients? Let me list a 1987 study by Najean along with his much more recent updates, and then a 1993 Italian study.
POLYCYTHEMIA VERA IN YOUNG PEOPLE: AN ANALYSIS OF 58 CASES DIAGNOSED BEFORE 40 YEARS by Najean, et al., publ. in British Journal of Haematology, vol. 67, 1987. The original published study included data on 58 patients (though only 46 were said to be clear PV - 12 were uncertain; 11 of these possible ET). Although I found the chart in the article rather confusing - only 16 patients shown evaluable at the 15th year; yet the data is listed here (and elsewhere) as over 70% of these young PVers surviving at the 15th year. (And, yes, this is what Najean says; it's just that the chart is confusing - to me, anyway!) The progression to spent phase/MF is 30% at the 15th year (or even higher - 37% if only the 46 clearly-diagnosed PVer's are included). (In older patients the authors note incidence of spent/MF to be 30% at 13 years; 50% at the 16th year.) The mean time to spent in phlebotomy-only is 9.5 years; in myelosuppressed patients it is 14 years. In this 1987 article, the authors recommend hydroxyurea for long-term treatment in hopes of reducing vascular accidents, but also in delaying evolution of the disease toward spent phase or MF.
When we turn to 1997 data reported by Najean (as in a letter to one of our members) and, briefly , in the new article published last month in BLOOD - we find that hydroxyurea has not substantially minimized the risk to spent/MF as the authors had hoped in 1987. In about 290 patients under 65 at diagnosis (99 under 50 years) with median age of 54, median survival ("median" meaning 50% living at this time) was 15.5 years for hydroxyurea (HU) and not yet obtained at 16 years for pipobroman (Vercyte). This is nearly identical to the PVSG study on HU though the median age there is higher. (Phebotomy-only has lower survival in PVSG studies than HU.) For those on HU, the risk of MF was 8% at the 8th year, 16% at the 10th; 24% at the 12th, 30% at the 14th, and over 40% at the 16th year. In those patients under 50 years at diagnosis; Najean states that the risk is still there and does not appear to be lower. He does feel strongly that young PV patients should be treated because of increased risk on phlebotomy-only. On the positive side, he states that 75% of young PV patients (under 50 when diagnosed) in the studies lived at least 16 years. Pipobroman having a higher survival than HU; leukemia nearly same (10% at 13th year) and spent/MF was considerably less.
POLYCYTHEMIA VERA AND ESSENTIAL THROMBOCYTHEMIA IN YOUNG PATIENTS By Frezzato, et al., publ. in Haematologica, vol. 78, 1993. This Italian study looked at 28 PV patients aged under 40 with mean follow-up of 8.2 years. Their survival rate was slighly lower than Najean's with 66% surviving at 15 years. The study does not really give us any break-down of survival as per treatment; thrombotic complications seem to be more of the emphasis. Spent phase is said to have occurred in only 1 of the PV patients, but note the characteristics for diagnosing the spent phase. Three of the following had to be present: increasing splenomegaly, teardrop RBC's, leukoerthroblastic blood picture, extensive bone marrow fibrosis.
To look at the PVSG longevity studies per treatment - there are reviews on our webpage. The articles are "The Very Long-Term Evolution of Polycythemia Vera: An Analysis of 318 Patients Initially Treated by Phlebotomy or 32P Between 1969 and 1981" and "From Efficacy to Safety: A Polycythemia Vera Study Group Report on Hydroxyurea in Patients With Polycythemia Vera." These articles are reviewed under the "From the Archives" section; look under "Seminars in Hematology." I was going to repeat some main points about those articles here as well, but this seems long enough!