Fw: Analysis on CML & Gleevec

[Guest guest]

Dear Group,

I found this article about Gleevec and secondary cancers. I am forwarding it

for everyone's benefit because I now realize that some of us are suffering from

secondary cancers, or have previously had secondary cancers. It was brought

even more to my attention while I was staying in an apartment in Houston for the

Ariad trial, where I met several people who were enduring secondary cancers.

Some of you have even lost

family members and friends in this way. This is data from Switzerland, but I'm

quite sure the percentages would be the same world wide. It is something that

looms over us and we should be aware of any different symptoms and bring up

anything unusual to our doctors, like a growing mole or a lump in a breast.

*****************************************

In our secondary analysis, we included all patients who were alive 2 years after

CML diagnosis and only second cancers which arose more than 2 years after CML

diagnosis. Person-years at risk were accumulated as above, but starting 2 years

after CML diagnosis.

Relative survival was calculated as the ratio between the observed survival of

the patient cohort (all causes of death included) and the expected survival of a

comparable group from the general population (assumed to be free of CML). This

provided a measure of total excess mortality associated with a diagnosis of CML

as compared with mortality in the general population, irrespective of whether

the excess mortality was directly or indirectly due to the original cancer.

Relative survival curves and their confidence intervals were estimated according

to the method of Ederer et al.. The derivation of the comparable group was

obtained by matching on sex, age (5-year intervals), and calendar period of

diagnosis (5-year intervals). Relative survival estimates in different calendar

periods of diagnosis were compared using a Poisson regression model (21).

Results:

A total of 2,801 patients diagnosed with primary CML between 1970 and 1995 were

included in the Swedish Cancer Registry. After exclusion of 48 patients

diagnosed before age 18 years, 2,753 patients were considered for analysis, of

whom 1,538 were males (56%). Table 1 shows the characteristics of this cohort

and of the subcohort of 1,302 patients who were alive and resident in Sweden 2

years after their CML diagnosis. Among the 2,753 patients considered for

analysis, 145 (5.3%) developed a subsequent cancer, 80 of which occurred 2 years

after the CML diagnosis.

The relative survival curves (Figure 1) showed improved survival for CML

patients after 1985 (P < 0.001), when interferon-a was broadly introduced as

front-line therapy. For example, 45% (95% CI: 43, 48) of the patients diagnosed

before 1985 and 60% (95% CI: 56, 63) of those diagnosed in 1985 or later

survived for at least 2 years after the diagnosis of CML.

Studies:

Therapy for CML has changed considerably in the last 30 years. Hydroxyurea or

busulfan was used before the introduction of interferon-a in 1985, and the first

tyrosine kinase inhibitor, imatinib, became commercially available in May 2001.

Our main aim in this work was to estimate the incidence of second malignant

neoplasms in CML patients treated without imatinib (CML diagnosis in 1970–1995).

This provides reference levels for any future work addressing the issue of

carcinogenic effects of chronic use of tyrosine kinase inhibitors in CML

patients. Although the Swedish national cancer registry has no information on

treatment, it is unlikely that the second malignancies analyzed here (diagnoses

made up to 1995) included cases that could be attributed to chronic use of

imatinib.

We found that, before the introduction of imatinib, CML patients had an 80%

relative increase in cancer incidence (SIR = 1.82, 95% CI: 1.53, 2.14) as

compared with the general population of Sweden, in both males and females,

patients diagnosed before and after 1985, and patients diagnosed before and

after age 50 years. Patients who had already experienced malignant tumors before

CML onset did not show a significantly higher risk of subsequent cancer,

probably because they did not survive very long (2-year relative survival = 34%,

95% CI: 28, 41).

Our finding of a significant increase in the incidence of urogenital tract

cancer challenges Roy et al.'s suggestion of a possible role of imatinib in

increasing the risk of this type of cancer. This increased risk was no longer

significant if prostate cancer alone was considered, but with the limited number

of these second cancers seen, the statistical power to detect a difference with

the general population was relatively low (approximately 30%). CML patients also

had a higher incidence of stomach cancer, which may deserve further

investigation in light of the association found by Miettinen et al. and the

possible link between hydroxyurea and stomach cancer. The observed increased

risk of skin cancer is consistent with observations on patients with all types

of myeloid leukemia. This study also confirms, in a European population, the

association with lymphoid leukemia found in the US Surveillance, Epidemiology,

and End Results registry. The increased risk of myeloid leukemia as a second

cancer is not reliable, since " blast crisis, " which was relatively common in CML

patients at that time, could easily have been registered as a second malignancy.

Relative survival curves showed that CML was related, in the pre-imatinib era,

to a very steep decline in survival. This was in spite of a marginal improvement

observed for patients diagnosed with CML after 1985, possibly related to

treatment with interferon-a and an increasing use of transplantation. It is

expected that treatment with tyrosine kinase inhibitors will dramatically

improve the survival of these patients.

Our findings have the strength of coming from a population-based study in which

CML was analyzed separately and not combined with other myeloid leukemias, as is

usually done in registry reports. This resulted in limited sample sizes for the

study and small numbers of second malignancies, due to the relative rarity of

CML. Thus, statistical power was limited, especially in subgroup analyses.

Although our data represent only the Swedish population, it is reasonable to

expect a similar pattern in other European countries.

The higher incidence of second malignancies in CML patients could be due to many

factors—for example, a genetic predisposition to cancer, exposure to risk

factors that are common to CML and other cancers, the CML treatment itself, or,

in the first years after diagnosis, increased medical surveillance. It was not

our aim in this study to explain causal relations; yet, for future studies with

chronic exposure to a tyrosine kinase inhibitor, an increased incidence of late

second malignancies (for instance, after 2 years of treatment) might generate

alarm regarding possible side effects. In this regard, it is worth noting that

according to our data from patients not treated with imatinib, SIRs for skin

cancer and stomach cancer increased among patients who survived beyond 2 years.

On the other hand, SIRs for urogenital tract cancers and lymphatic leukemia

decreased after 2 years from diagnosis.

In summary, these estimates of the incidence of second malignancies in CML

patients, as well as registry-based long-term survival figures, constitute a

useful relevant reference for future studies and a benchmark for comparison with

estimates obtained in the era of tyrosine kinase inhibitors.

http://www.medscape.com/viewarticle/734061_4

********************

FYI,

Lottie Duthu