SHIVA: Trial of Personalized Cancer Care Disappoints

Alexander M. Castellino, PhD

September 23, 2015

The idea of personalized cancer care based on molecular characteristics of the tumor promises to expand the boundaries of precision medicine. Numerous case reports and other observations have suggested that therapy targeted at molecular chartacteristics of a tumor can have substantial effects.

However, the first randomized trial to compare this approach with conventional therapy has yielded rather disappointing results. There was no difference in progression-free survival (PFS) between the two treatment groups.

The results from this trial, known as SHIVA, were published online September 3 in Lancet Oncology. The study was also presented earlier this year at the American Society of Clinical Oncology (ASCO) 2015 annual meeting.

The trial was conducted in 195 patients with metastatic cancer (any solid tumor) refractory to standard of care. The patients were randomly assigned to receive either molecularly targeted agents (used off label) chosen on the basis of the molecular profile of the tumor or therapy based on the physician's choice. The median follow-up period was 11.3 months.

The results showed a median PFS of 2.3 months for patients receiving targeted therapy vs 2.0 months for patients receiving therapy based on the physician's choice.

"So far, no evidence from randomised clinical trial supports the use of molecularly targeted agents outside their indications on the basis of tumour molecular profiling," write the SHIVA investigators, led by corresponding author and principal investigator Christophe Le Tourneau, MD, PhD, of the Department of Medical Oncology, Institut Curie, Paris, France.

"Our findings suggest that off-label use of molecularly targeted agents outside their indications should be discouraged, and enrolment into clinical trials encouraged," Dr Le Tourneau and colleagues add.

"SHIVA is a notable trial. It is the first to test, with a randomised control, the idea of whether off-label use of commercial drugs for matched molecular biomarkers confers a clinical benefit," writes Daniel V. T. Catenacci, MD, in an accompanying editorial.

Dr Catenacci, from the Department of Medicine, Section of Hematology/Oncology, at the University of Chicago, in Illinois, told Medscape Medical News that the approach of matching a drug to a target is generally based on case study reports, observational cohorts, and meta-analyses — all without appropriate controls.

"This prospective randomised trial was negative, a common occurrence after the publication of several promising uncontrolled reports," Dr Catenacci writes in his editorial.

Richard L. Schilsky, MD, chief medical officer of ASCO, who is not associated with the SHIVA study, told Medscape Medical News: "Although negative, this is an important study and the first randomized one done in the complex area of matching drugs to genomic profiles of tumors."

"What SHIVA tells us is that we need to continue doing research in this area of personalized medicine," he added. He echoed sentiments expressed by the investigators and editorialist. "One of the most important take-aways from this study is that the practice of matching drugs to mutational profiling of tumors should not be done outside of a research study."

The SHIVA Study

The SHIVA study was a proof-of-concept, open-label, randomized, controlled study carried out at eight academic centers in France.

Patients were eligible for the study if they had recurrent or metastatic solid tumor and had failed standard of care — any treatment regimen with known survival or quality-of-life benefit as determined in randomized trials.

Single-agent treatment was assigned to enrolled patients on the basis of molecular profiles affecting known targetable pathways (eg, hormone receptor, P13K/AKT/mTOR, and RAF/MEK pathways) and included agents that were approved for clinical use in France but were used outside of their indications (erlotinib, sorafenib, imatinib, dasatinib, vemurafenib, everolimus, abiraterone, letrozole, and tamoxifen).

SHIVA was a histology-agnostic study, in which tumor histology was not taken into account when treatments were assigned.

If tumors had several targetable mutations, prioritization of therapy was provided on the basis of criteria established by the Molecular Biology Board.

Between 2012 and 2014, 741 patients underwent screening. Complete tumor profiles were available for 496 patients; no targetable molecular alteration was available for 203 patients; 293 pateints were enrolled in the study, and 195 patients were randomly assigned. Of these, 99 patients received a targeted agent, and 96 patients were treated on the basis of physician's choice of therapy ― they received single-agent treatment (76%), combination therapy (21%), or best supportive care (3%).

The study objective was to detect a 40% longer PFS compared with the control group.

Study Results

Baseline patient characteristics were similar across the two groups with respect to sex, previous lines of therapy, and molecular pathways altered.

Of the 195 patients who underwent randomization, 55 had two or more molecular alterations; 82 had a molecular alteration affecting the hormone receptor pathway, 89 had alterations in the PI3K/AKT/mTOR pathway, and 24 had alterations in the RAF/MEK pathway.

The hazard ratio for disease progression was 0.88 (95% confidence interval, 0.65 - 1.19; P = 0.41) and was not significant.

Explaining SHIVA and Going Forward

Dr Le Tourneau and his colleagues provided some insights into why SHIVA was a failed study.

First, variability in analysis associated with the use of multiple treatment groups with multiple molecular alterations across tumor types and histologies may have contributed to SHIVA's disappointing results. The treatment algorithm used was unidirectional, with a single molecular alteration empowered to predict the efficacy of molecular targeted agents.

"Whether multidimensional treatment algorithms that incorporate information from several genes using systems biology approaches will be able to better predict response to molecularly targeted agents remains to be established," the SHIVA investigators write in their discussion.

In SHIVA, molecularly targeted agents were used as single agents. This approach may well be associated with reduced efficacy, owing to the development of resistance, the investigators suggest. "Use of several molecularly targeted agents in combination is an appealing way to counteract resistance," the investigators write.

In addition, spatial and temporal intrapatient tumor heterogeneity was not considered. "Improvements in molecular analyses of circulating tumour DNA might allow tumour evolution to be readdressed over time and therapy potentially revised accordingly," the SHIVA investigators propose.

Finally, only drugs marketed in France were used in the study.

"On the surface, SHIVA is a setback," Dr Catenacci told Medscape Medical News. "But there are nuances that must be taken into account — specific drugs were used; specific tumor types were tested; and specific biomarker classes were targeted," he said.

He writes in his editorial: "[C]areful consideration of the variables embedded in this trial design is warranted. These include choice of the biomarker groups, the molecular profiling assays and positivity criteria, the drugs, the treatment assignment algorithm, and the histology makeup, all of which contributed to the aggregated results of SHIVA."

Because of the variables introduced from this treatment strategy, the conclusions from SHIVA cannot be generalized, Dr Catenacci suggested.

"But this fact does suggest that any other proposed strategies should be similarly tested before they are accepted as routine standard care," he added.

Other proposed strategies should be similarly tested before they are accepted as routine standard care. Dr Daniel Catenacci

"This study is one of several (completed, ongoing, proposed) that seek to further define and refine the role of targeted sequencing of tumors to guide and improve therapy. The entire enterprise depends on accurate testing and active agents targeting functional pathways," medical oncologist Clifford A. Hudis, MD, told Medscape Medical News.

Dr Hudis is vice president for government relations and chief advocacy officer and chief, Breast Medicine Service, at Memorial Sloan Kettering Cancer Center, in New York City. He was not associated with SHIVA.

"The landscape now includes notable successes and some disappointments, but the approach here ― of testing the overall strategy as opposed to testing specific agents and a specific target ― may make it harder to identify promising approaches," Dr Hudis added.

Dr Catenacci goes further. In his editorial, he said: "[N]ext-generation regulation must accompany next-generation trials through the integration of next-generation companion diagnostics and the concept of personalised treatment strategies, to continue to advance clinical cancer care."

In the context of next-generation trials, the ASCO's TAPUR trial and the National Cancer Institute's MATCH trial will evaluate between 10 and 15 drugs contributed by five pharmaceutical companies in cohorts of up to 25 patients grouped in accordance with tumor type and genetic abnormality. Each drug will be matched with its appropriate target.

These trials are different in design from SHIVA. "TAPUR and MATCH are not designed to prove a strategy but to answer the question of whether the drugs tested can generate an activity signal in that setting," Dr Schilsky, who is also heading ASCO's TAPUR study, told Medscape Medical News.

Dr Schilsky explained that the rules of engagement are prespecified. "A set of matching rules match drugs with corresponding genomic abnormalities," he said.

"These are also histology-agnostic studies," Dr Catenacci told Medscape Medical News. TAPUR and MATCH are nonrandomized studies and are matching a drug with the molecular profile of the tumor.

"There is much tumor heterogeneity in these studies since they are being studied across many tumors. With respect to the drugs, we will not know until the trials are complete whether the drug is tumor specific or not," Dr Catenacci added.

One of the authors of the SHIVA study reported receiving grants from multiple pharmaceutical companies. Dr Catenacci, Dr Schilsky, and Dr Hudis have reported no relevant financial relationships.

Lancet Oncol. Published online September 3, 2015. Abstract, Editorial


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.