(3) PBSC Infusions: 
(a) PBSC collections contain approximately 2-300 ml of volume. Immunologic 
reactions are impossible in identical twins. However a high concentration 
of granulocytes could lead to pulmonary sequestration resulting in a low 
concentration of oxygen. Patients will be closely observed for such 
untoward reactions. 
(4) Engraftment: 
(a) Failure to engraft will be defined as not achieving a granulocyte level of .5 
X 10 9 /L by day 28 and/or failure to achieve a platelet count of 20 X 10 9 /L 
without transfusions by day 36. Marrow from the identical twin will be 
aspirated and infused at this point. Late graft failure is defined as 
achieving engraftment of granulocytes and platelets with a subsequent loss 
of peripheral counts defined as a granulocyte level < .5 X 10 9 and/ or 
platelets < 20 X 10 9 requiring platelet transfusions. Marrow from the 
identical twin will be aspirated and infused at this point irrespective of the 
etiology. 
11. Records. 
Records will be maintained on this study by Kathy Lilleby. 
12. Statistical Considerations. 
In review of engraftment data for 234 autologous transplant patients with diseases other than 
AML, 16 (7%) had granulocytes counts of 0 after day 19, and 26 (12%) of the 220 who reached 
500 granulocytes did so after day 33. Based on the notion that these patients represent outliers, 
these criteria will be employed as a working definition of primary graft failure for purposes of 
this protocol. Initially we plan to enroll 10 patients. If any two patients have primary graft 
failure, the trial will be closed. This algorithm assures a 74% chance of enrolling 10 consecutive 
patients if the true risk of graft failure is less than 10% and an 80% chance of not enrolling 10 
patients if the risk exceeds 30%. The review of engraftment data also showed that among 
autologous patients with diseases other than AML, 14% of those whose granulocyte counts 
surpassed 500 subsequently had a decrease to less than 500. Of these, 20% died without again 
reaching granulocyte counts of 500. Thus approximately 3% of all patients died with secondary 
graft failure. The present study will be closed if any engrafted patient (granulocytes greater than 
500 for three consecutive days) subsequently dies with a granulocyte count less than 500 without 
an obvious explanation such as relapse, CMV infection or drug toxicity. This algorithm assures a 
78% chance of enrolling 10 patients if the true risk of death from secondary graft failure is less 
than 2.5% and a 79% chance of not enrolling 10 patients if the risk exceeds 16%. 
The secondary endpoints of this study are time to achieve a self-sustaining granulocyte levels of 
100, 200, 500 and 1000/mm 3 and untransfused platelet levels of 20,000, 50,000 and 100,000/mm 3 . 
Other endpoints will include the number of platelet and red cell transfusions administered. 
These endpoints will be compared to 3 historical control groups: 1. identical twins receiving 
marrow alone, 2. identical twins receiving marrow and post-transplant growth factors and 3. 
identical twins receiving marrow plus PBSC’s. 
Morbidity and expense of PBSC collections versus marrow aspiration will be determined. - 
Recombinant DNA Research, Volume 16 
[369] 
