Activity of Single Agent Drugs in Phase I and II Trials Best Predicts Future Clinical Utility in Multiple Myeloma
Bibliographic record
Abstract
Abstract Abstract 4034 Through literature review we identified 376 pre-clinical studies in which a small molecule or antibody has been demonstrated to be cytotoxic to multiple myeloma (MM) cells either as cell lines, primary tumor samples or in mouse models. Indeed it seems there is no limit to the number of compounds that are cytotoxic to MM cell lines at high enough concentration. Since the pipeline to clinical trials is then large the utility of such assays in predicting future clinical value is debatable. We therefore sought to examine the utility of applying very early Phase I or Phase II trial experiences to predict for future clinical adoption. Using Pubmed, ASCO, ASH, EHA and IMWG abstracts we identified 117 drugs explored as single agents in 211 early stage clinical MM trials between 1961 and 2012 involving 7,166 patients. As definitions have changed over time, our analysis used partial responses (PR) or better as a response. When multiple trials were reported with a single drug, the average and the highest response rate for that agent was identified. Results were considered regardless of whether single or multi center trial, elderly or younger patients, stage of disease, targeted or non targeted therapy. High dose therapies requiring autologous stem cell support (eg. melphalan and etoposide), trials with less then 5 patients (e.g. siltuximab) or trials with concomitant steroid application which obscure single agent activity (e.g. pomalidomide, marizomib or daratumumab) were excluded in the analysis. Results: 17 drugs (14.5%) showed a best result of PR in more than 20% of patients, but when average response across all reported trials is considered, doxorubicin, arsenic trioxide and idarubicin drop out leaving 14 active drugs. Twelve drugs (10.3%) report as a best result PR in 10–20% of patients. 88 drugs (75.2%) reported a PR response in less than 10% of patients, 68 of which (58.1%) had no reported activity. A striking finding of our study is that all drugs approved for, or commonly used in clinical practice (prednisone, dexamethasone, cyclophosphamide, melphalan, bendamustine, thalidomide, lenalidomide, bortezomib, carfilzomib) have an average single agent PR response of at least 20%. Pomalidomide, although not approved, meets this threshold in Phase I testing. Older agents including teniposide, fotemustine, paclitaxel, and interferon also appear active by this criteria but have fallen from favor or were never adopted due to perceived toxicity. Of note, a number of newer agents which are in, or which recently completed Phase 3 testing, such MLN9708, elotuzumab, panobinostat, vorinostat or perifosine show little single agent activity. A weakness in this approach is that changes in measurement of response over time might bias the results, especially in favor of older studies. Furthermore, the chance of responding to a new drug has likely slowly declined as therapies have improved and patients have entered such trials at increasingly late stages of disease. It also does not rule out the possibility of synergistic activity or clinical benefit from a cytostatic drug. Nevertheless, our analysis suggests that a cut off of 20% single agent partial response activity averaged across all trials with that agent is highly predictive of future clinical success. If best reported response over 20% is considered no drug except vincristine has yet reached the clinic and has subsequently fallen from favor. Thus only drugs with 20% PR activity are in widespread use and thus this benchmark provides a framework for guiding choice of drugs for late stage clinical testing. Disclosures: Jimenez-Zepeda: MMRF: Research Funding; Jansenn Ortho: Honoraria. Fonseca:Onyx: Consultancy, Research Funding; Cylene: Research Funding; Medtronic: Consultancy; Otsuka: Consultancy; Celgene: Consultancy; Genzyme: Consultancy; BMS: Consultancy; Lilly: Consultancy; Binding Site: Consultancy; Millenium: Consultancy; AMGEN: Consultancy; Mayo Clinic: Patents & Royalties. Stewart:Millenium: Consultancy, Honoraria, Research Funding; Onyx: Consultancy; Celgene: Consultancy.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.012 | 0.045 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.004 | 0.004 |
| Bibliometrics | 0.011 | 0.013 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".