ALL Leukemia

SLIDES & TRANSCRIPTS
Monday, May 12, 2003

Adolescent ALL

Sima Jeha, M.D.

Slide 1:

Good morning. There has been a lot of debate over the past few years on what defines childhood, when does it end, where does adulthood begin, what defines adolescence, who should treat adolescents with TLL, and on what protocol.

Some people think the answer is very easy. ALL is the most common malignancy in pediatrics. Therefore, the pediatric oncologists are experts in the disease. This is reflected by the very good improvement and outcome that has occurred over the years.

Also, the pediatric oncologist's practice and referring centers have a lot of support and resources, and the ratio of physicians to patients really allows very close follow up by the physician to the patient.

All of this is correct. However, I would like to point out that our colleagues in adult medicine have also made major improvements over the past few years, but they are dealing with a more difficult disease, it seems.

Slide 2:

Actually, if we look at the hyper CIVE and we take only the standard risk, the good risk adult patients -- and those are all adult patients, not just young adults and adolescents -- we see that they constitute only 25 percent of the total adults at Anderson.

You notice that the event free survival is very close to the one that is reported on the high risk pediatric protocols.

Slide 3:

Now, if we look at the pediatric data, and we have in blue the CCG outcome and red the other outcome by SSER, I will exclude the infants, who do not do well, and I will only focus on over one year of age.

As we know, the one to nine years do very well on standard therapy with minimal chemotherapy, minimal intensive chemotherapy.

Now, over 10 years, we intensify the pediatric protocol and we give more chemo, more intensive chemo. Still, the outcome starts dropping over 10. Interestingly, even it drops at over 14, even on the CCG studies that are the best pediatric studies for ALL.

Slide 4:

This is reflected on all other pediatric studies. As you notice, there is a difference in outcome that is expected on different studies at different centers.

However, what is very common among all the studies, with the CCG POG and St. Jude or RFM, is that the kids over 10 years do not do as well as the kids that are younger than 10 years of age.

Slide 5:

If we look at adult data, the same trend in effect of age seems to apply, too, and it seems that adults that are younger than 30 do much better than adults over 30 or over 40.

I would like to point that on the hyper CVAD, if you are under 30, the CR rate is 98 percent, which is very comparable to the one we have in pediatrics, and the percent survival is 54. This going up all the way to 30.

Slide 6:

So, what causes this? Of course, there is a difference in referral and doctors, but also since we have differences in the same centers treating the same populations, what causes that difference?

We know there is more Philadelphia positive translocation in the adult population compared to pediatrics. We know that there is much more hyperdiploidy in pediatrics compared to adults. So, there is more high risk in the adult population compared to pediatric.

Also, the pediatric patients tolerate chemotherapy better. There are studies that indicate they are more sensitive to methotrexate, vincristine, prednisone and asparaginase, which are the backbone of treatment in pediatrics.

One last thing is the treatment. On all the studies, when we compare them, although the variables are many, including the centers and the doctors, what is very common is that pediatricians use more vincristine, more asparaginase and more steroids, and this is really a common theme on every study trying to compare both arms.

Slide 7:

Now, I want to show this slide to show you that it gets more complicated. Even when you look at the outcome of a subtype of leukemia treated by the same group of patients -- in this case it is Philadelphia positive ALL that is treated with pediatricians -- the outcome is 28 percent, it is higher than adults. So, even with the bad prognosis, the Philadelphia positives, the pediatrics do better than adults.

Now, even if you take the pediatrics and stratify them, over 10 and less than 10, and white count over 25, the patients who are younger than 10 years old, and who have the white cell count of less than 25,000, their event-free survival goes over 50 percent.

The ones that are over 10, and who have the white count over 25,000, have a lower survival than the 30 percent. So, there is something more to it, even when we look at the subtypes.

Also, to stress the importance of the subtype on prognosis --

Slide 8:

and this is the slide that Mitch was lacking, probably -- it is the result of B cell mature ALL on hyper CIVE.

You notice that he did show that the survival was close to the 50 percent. However, if we look at the age, even if we go as high as 60, the survival rate was 77 percent, which was really very close to what the pediatric studies show.

Slide 9:

I will show you two major studies that have compared head to head the same group of patients, age group, treated by pediatricians or by adult oncologists.

This study most of you are familiar with. It was presented at ASH three years ago. It compared patients treated on CCG versus patients treated on CALGB.

You notice that there were more unfavorable cytogenetics on the CALGB. There are more studies ongoing now to publish those results, and to look at where the median age was on each studies, and to look at other characteristics including how much of the treatment was delivered, and about whether the compliance was the same in both groups.

However, you notice that the CR rate is higher on the CCG protocol, 96 percent, versus 93 percent on CALGB. Also, the six-year event-free survival is higher on the CCG.

Slide 10:

Now, this was recently published in JCO, and it is comparing two French protocols, the pediatric FRALLE and the adult LALA.

The age group was 15 to 20, and I notice here the median age was 16 on the FRALLE and it was 19 on the LALA. All other characteristics were equal between the patients.

You notice that the CR rate is 83 percent only on the LALA. So, the difference between the FRALLE and the LALA as far as CR, the band of induction, is even higher than between the CLGB and CCG.

The difference in the induction was mostly that the FRALLE used much more higher dose induction than the LALA. They also used asparaginase, which was not used n the LALA.

The difference continued, even beyond induction. Also, they looked at the compliance and giving chemo on time. The patients on FRALLE got the chemo on time, without delays, whereas on the LALA there were more delays.

Slide 11:

I tried to look at all the variances between centers and groups. So, I tried to look at M.D. Anderson, at patients between the ages of 10 and 19.

Pediatrics means they were treated on the pediatrics protocol and adults means they were treated on adult protocols. So, this is the difference in service and in protocol.

You notice that only one patient was 13, and in pediatrics most of the 10 to 19 were under 16 and almost divided between 16 and 13. We didn't have anybody over 16, whereas the majority of patients on the adult services were over 16. So, we have a difference in age.

Slide 12:

When we looked at the characteristics of the 13 to 16, this is the group I wanted to compare head to head, because we really have overlap. The characteristics were very similar.

Slide 13:

Now, when we looked at the curves, it was very interesting. These are the pediatric curves, and it compared less than 10 years. It compares with what we have nationwide.

Now, if we look at the 10 to 19, treated on the adult service, it is the green line, and it looks worse than the 20 to 39 on the adult service, whereas the 10 to 19 on pediatric protocol looks much better.

So, I went back to see the reasons for that.

Slide 14:

It became even more interesting. Those are the 10 to 12 year olds treated on the pediatric protocol. The green is the 13 to 16 treated on pediatric protocol. Now, those are the 13 to 16 treated on the adult protocol.

They not only do much worse than the pediatric patients, but also worse than the other patients treated on the adult protocol. So, I decided there was something funky with this group.

Slide 15:

When I looked more, I noticed something that would teach us that every report should be examined really carefully.

I am not sure what happened here, but the patients who are 15 to 16, whether they are treated on pediatric or adult protocol, for some reason, did very poorly at M.D. Anderson in the period studied.

The pediatric patients had good curves and the adult patients only had one, who was still alive, and of course he was not counted on this.

I am not sure what the reason for this is. I am looking into it now, but this is to show you how a few can change a curve, and people have to be very careful when comparing the results.

Slide 16:

Why is this very important, discussing the teens? It is because of the 2,500 children with ALL diagnosed every year and 2,000 adults.

Most of the children are young, do well on standard protocol, and it is not a problem. The other majority of patients are over 40, because that is the second peak.

So, we have in between patients who have an intermediate prognosis, between 80 percent and 30, and they get treated on different protocols.

So, the best way for us to really understand the role of biology and age, in my opinion, would be to treat them on the same protocol, especially in this age of targeted chemo, where we will start dissecting more the protocol according to phenotypes and genotypes.

I think the idea of CCG, CALGB treating the patients on the same protocol is great, because it will give us an idea how the groups fared when they treat the patients on the same protocol.

I feel that at centers like Anderson, with our in-house protocol, probably patients between the ages of 10 and 40 should be also treated on the same protocols, since they can validate chemotherapy, intensive chemotherapy, and since the prognosis is not as good as children.

This way we will decrease the variables and we will be able to really deduct conclusions about what causes the poorer outcomes in adolescents and young adults. Thank you.