Precision medicine: Lessons learned from the SHIVA trial

TY - JOUR

T1 - Precision medicine

T2 - Lessons learned from the SHIVA trial

AU - Tsimberidou, Apostolia M.

AU - Kurzrock, Razelle

N1 - Funding Information: Apostolia M Tsimberidou a atsimber@mdanderson.org Razelle Kurzrock b a Department of Investigational Cancer Therapeutics, The University of Texas MD Anderson Cancer Center, Houston, TX, USA Department of Investigational Cancer Therapeutics The University of Texas MD Anderson Cancer Center Houston TX USA b Center for Personalized Cancer Therapy, and Division of Hematology and Oncology, UCSD Moores Cancer Center, La Jolla, CA, USA Center for Personalized Cancer Therapy and Division of Hematology and Oncology UCSD Moores Cancer Center La Jolla CA USA The SHIVA study 1 is the first randomised investigation of precision therapy. Its findings contradict published data. In our opinion, the study has serious weaknesses. First, patients with cancer often have several molecular alterations, and many patients are unlikely to respond to monotherapy. 2 In patients with PI3K/Akt/mTOR pathway abnormalities, we showed significantly less frequent responses with monotherapy versus combinations. 2 Second, everolimus weakly affects the PI3K/Akt/mTOR pathway; patients with PI3K/Akt/mTOR alterations often have coexisting alterations in RAF/MEK that confer resistance. Patients with RICTOR alterations, which would activate the mTOR2 complex, were to be matched with everolimus (an mTOR1 complex inhibitor), despite the possibility that mTOR1 inhibition in this situation could hyperactivate the pathway. Third, many patients had hormone receptor abnormalities; response to hormone monotherapy in pretreated patients with advanced disease is unlikely. 3 Fourth, some matches were incorrect: imatinib, an ineffective RET inhibitor (IC 50 37 μmol/L) was matched to RET alterations. 4 Fifth, precision medicine is defined by the presence of biological data showing that a targeted drug affects a molecular alteration and needs to be taken in the context of clinical experience—an unmet criterion for many matches in SHIVA. Finally, patients randomly assigned to targeted therapy were treated using a predefined algorithm by contrast with the control group who were assigned therapy by the physician. Because physicians could take into consideration comorbidities in the control group, this design could have conceivably introduced bias. So far, several studies support the use of precision medicine: Von Hoff and colleagues, 5 using patients' prior therapy progression-free survival (PFS1) as a control for the matched therapy progression-free survival (PFS2), reported that 27% of matched patients had a PFS2:PFS1 ratio of 1·3 or more. We have shown that matched therapy was associated with a higher proportion of patients achieving a response and longer survival and time to treatment failure than unmatched therapy, and with a longer time to treatment failure than patients' prior systemic therapy. 6 Matched therapy was also an independent factor predicting outcomes in multivariate analyses. 6 Precision medicine for patients with pretreated breast cancer 7 resulted in 44% of patients having a PFS2:PFS1 ratio of 1·3 or more. A multisite study 8 (with >1000 patients with lung cancer) showed that matching was associated with longer survival than that seen in patients without genotype-directed treatment, while a study 9 of vemurafenib in patients with BRAF mutations showed responses in multiple cancer types. Two meta-analyses 10,11 in 70 000 patients reported that trials with a “personalized” strategy led to a higher proportion of patients achieving responses, and longer progression-free and overall survival than trials with unselected patients. Therefore, the conclusion from the SHIVA study should not be that precision therapy is disappointing and that the use of targeted drugs off-label should be discouraged, but that hormone monotherapy or weak targeted drugs, especially as single agents, are ineffective in advanced cancer. In view of the substantial data suggesting the positive anticancer activity of targeted therapy, including off-label drug use, on clinical outcomes in appropriately selected patients, this Article highlights the principle that, to implement precision medicine, a targeted drug should effectively and consistently affect the function of a driver alteration and combination regimens need to be used for many patients with advanced disease. Results of continuing studies across tumour types—eg, our randomised Initiative for Molecular Profiling and Advanced Cancer Therapy (IMPACT) 2 trial, ASCO's Targeted Agent and Profiling Utilization Registry (TAPUR) trial, and others—will further elucidate the value of precision medicine. AMT receives research funds from Foundation Medicine, EMD Serono, Baxter, Onyx Pharmaceuticals, and Bayer. RK receives consultant fees from Sequenom Inc, has an ownership interest in RScueRx Inc, and receives research funds from Genentech, Pfizer, Merck Serono, Sequenom, Guardant, and Foundation Medicine.

PY - 2015

Y1 - 2015

UR - http://www.scopus.com/inward/record.url?scp=84961662117&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84961662117&partnerID=8YFLogxK

U2 - 10.1016/S1470-2045(15)00397-6

DO - 10.1016/S1470-2045(15)00397-6

M3 - Letter

C2 - 26678197

AN - SCOPUS:84961662117

SN - 1470-2045

VL - 16

SP - e579-e580

JO - The lancet oncology

JF - The lancet oncology

IS - 16

ER -