Additonal info on MDX-1106 OR BMS-936558

On May 20th I blogged my on take on the abstract of an article talking about the BMS version of anti-PD1.  The article was finally published on June 2, 2012 in the New England Journal of Medicine...Safety, Activity, and Immune Correlates of Anti-Pd1 Antibody in Cancer.  Written by the same cast of characters already noted.

Though the final article has slightly different numbers (due to the final accounting of patients and their results, rather than the arbitrary cut-off in the abstract)....Bottom line: 296 patients with a variety of cancers (94 with melanoma) were treated every 2 weeks for for up to 2 years at dosages of 0.1 to 10 mg/kg. Patients with non-small-cell lung cancer, melanoma, and renal-cell cancer demonstrated some response.  Patients with prostate and colon cancer did not.  All these patients were very ill and "the majority of patients were heavily pretreated; 47% had received at least three prior treatment regimens."  A maximum dose was not defined and researchers deemed the drug well tolerated (PER RESEARCHERS!!!!!  though the dead people and those suffering with side effects may not agree!!!!!), though "15 of the 296 patients discontinued treatment owing to treatment related adverse events....As of the date of analysis, 62 patients (21%) had died; disease progression was the most common cause of death....[though] there were three drug-related deaths (1%) due to pneumonitis..."  Some patients at all drug dosage levels experienced a response, indicating that as yet...we do not know the most effective dose.  Response seems to be durable for some patients....at times lasting greater than one year at publication.

However, perhaps the biggest news from the study/article is the significance of the presence of PD-L1 in the tumors of patients treated.  "61...tumor specimens from 42 patients (18 melanoma, 10 lung cancer, 7 colorectal cancer, 5 renal cell, and 2 with prostate cancer) were analyzed for  PD-L1 expression on the surface of tumor cells."  25 of the 42 patients had tumors that were positive for PD-L1. "Of these 25 patients, 9 (36%) had an objective response.  NONE of the 17 patients with PD-L1 negative tumors had an objective response.....These preliminary results must...be interpreted with caution."  However, the presence (or absence) of such bio-markers may play a huge role in the identification of candidates for anti-PD1 treatment and guide disease management decisions.

So, it would seem that if you have PD-L1 on the surface of your tumor....you have a 36% chance of a response to the BMS anti-PD1 product.  That number is better than the 30% chance that is showing up in most of the anti-PD1 data (limited though it is) thus far.  On the flip side, if you do not have that particular ligand...it appears that it is extremely unlikely that anti-PD1 therapy will help you at all.  To that end...BMS is taking this info very seriously.  They are paying for all the patients at Moffitt, in my study, to have their tumors tested for that very biomarker.  You do not have to agree to this testing to remain in my study, but they are requesting it of all of us, and I signed the agreement for mine to be tested Friday.  For those of us in the study, it is of no consequence.  We either have it or we don't.  We have already been treated and we will respond or we won't.  But, perhaps our data will help others.  First:  We need to see if this data holds.  Second:  If it does, perhaps it will help others make better treatment choices.  If this is true, and anti-PD1 cannot help you if your tumor does not express PD-L1, there is no need for you to waste precious time, money, and risk significant side effects seeking this particular treatment.  Obviously, time will tell....hopefully progress and knowledge will be attained soon. - c