Optimizing One-Sample Tests for Proportions in Single- and Two-Stage Oncology Trials
Simple Summary
Abstract
1. Introduction
2. Convolution Estimator
2.1. Toy Example
2.2. Convolution Approach and Exact Binomial Test Comparison
3. Two Stage Design
Convolution Approach and Simon Two-Stage Design Comparison
4. Real World Examples
4.1. One-Stage Designs
4.1.1. Example 1
4.1.2. Example 2
4.2. Two-Stage Designs
4.2.1. Example 1
- For , : Assuming 7 out of 27 responses, the futility p-value was 0.011, below the stopping threshold . The final-stage p-value, assuming 8 out of 35 total responses, was 0.0158, significant at the adjusted level .
- For , : Assuming 6 out of 24 responses, the futility p-value was 0.002, below . The final-stage p-value based on 8 of 36 responses was 0.0162, also significant at .
- For , : Assuming 5 out of 18 responses, the futility p-value was 0.015, which is less than . The final p-value with 8 of 37 responses was 0.020, significant at .
4.2.2. Example 2
5. Conclusions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- Prepared by Battelle Technology Partnership Practice. Biopharmaceutical Industry-Sponsored Clinical Trials: Impact on State Economies. Prepared for Pharmaceutical Research and Manufacturers of America (PhRMA). 2015. Available online: http://orclinicalresearch.com/wp-content/uploads/2018/12/battelle-2015-study.pdf (accessed on 8 July 2025).
- Leighl, N.B.; Nirmalakumar, S.; Ezeife, D.A.; Gyawali, B. An Arm and a Leg: The Rising Cost of Cancer Drugs and Impact on Access. Am. Soc. Clin. Oncol. Educ. Book. Am. Soc. Clin. Oncol. Annu. Meet. 2021, 41, e1–e12. [Google Scholar] [CrossRef] [PubMed]
- Kapinos, K.A.; Hu, E.; Trivedi, J.; Geethakumari, P.R.; Kansagra, A. Cost-Effectiveness Analysis of CAR T-Cell Therapies vs. Antibody Drug Conjugates for Patients with Advanced Multiple Myeloma. Cancer Control 2023, 30, 1–8. [Google Scholar] [CrossRef] [PubMed]
- Hoover, A.; Reimche, P.; Watson, D.; Tanner, L.; Gilchrist, L.; Finch, M.; Messinger, Y.H.; Turcotte, L.M. Healthcare cost and utilization for chimeric antigen receptor (CAR) T-cell therapy in the treatment of pediatric acute lymphoblastic leukemia: A commercial insurance claims database analysis. Cancer Rep. 2024. Epub ahead of print. [Google Scholar] [CrossRef] [PubMed]
- Simon, R. Optimal two-stage designs for phase II clinical trials. Control. Clin. Trials 1989, 10, 1–10. [Google Scholar] [CrossRef] [PubMed]
- Hamaker, H.C.; van Strik, R. The Efficiency of Double Sampling for Attributes. J. Am. Stat. Assoc. 1955, 50, 830–849. [Google Scholar] [CrossRef]
- Duncan, A.J. Quality Control and Industrial Statistics; Irwin: Homewood, IL, USA, 1986. [Google Scholar]
- Hutson, A.D. Modifying the Exact Test for a Binomial Proportion and Comparisons with Other Approaches. J. Appl. Stat. 2006, 33, 679–690. [Google Scholar] [CrossRef]
- Chernick, M.R.; Christine, Y.; Liu, C.Y. The Saw-Toothed Behavior of Power Versus Sample Size and Software Solutions: Single Binomial Proportion Using Exact Methods. Am. Stat. 2002, 56, 149–155. [Google Scholar] [CrossRef]
- Whitlock, M.C. Combining probability from independent tests: The weighted Z-method is superior to Fisher’s approach. J. Evol. Biol. 2005, 18, 1368–1373. [Google Scholar] [CrossRef] [PubMed]
- Asahina, H.; Oizumi, S.; Takamura, K.; Harada, T.; Harada, M.; Yokouchi, H.; Kanazawa, K.; Fujita, Y.; Kojima, T.; Sugaya, F.; et al. A prospective phase II study of carboplatin and nab-paclitaxel in patients with advanced non-small cell lung cancer and concomitant interstitial lung disease (HOT1302). Lung Cancer 2019, 138, 65–71. [Google Scholar] [CrossRef] [PubMed]
- Boudadi, K.; Suzman, D.L.; Anagnostou, V.; Fu, W.; Luber, B.; Wang, H.; Niknafs, N.; White, J.R.; Silberstein, J.L.; Sullivan, R.; et al. Ipilimumab plus nivolumab and DNA-repair defects in AR-V7-expressing metastatic prostate cancer. Oncotarget 2018, 9, 28561–28571. [Google Scholar] [CrossRef] [PubMed]
- Gridelli, C.; Perrone, F.; Gallo, C.; De Marinis, F.; Ianniello, G.; Cigolari, S.; Cariello, S.; Di Costanzo, F.; D’Aprile, M.; Rossi, A.; et al. Vinorelbine is well tolerated and active in the treatment of elderly patients with advanced non-small cell lung cancer. A two-stage phase II study. Eur. J. Cancer 1997, 33, 392–397. [Google Scholar] [CrossRef] [PubMed]
- Sundquist, F.; Georgantzi, K.; Jarvis, K.B.; Brok, J.; Koskenvuo, M.; Rascon, J.; van Noesel, M.; Grybäck, P.; Nilsson, J.; Braat, A.; et al. A Phase II Trial of a Personalized, Dose-Intense Administration Schedule of 177Lutetium-DOTATATE in Children With Primary Refractory or Relapsed High-Risk Neuroblastoma-LuDO-N. Front. Pediatr. 2022, 10, 836230. [Google Scholar] [CrossRef] [PubMed]
Run | y | x | Convoution p-Value | Exact Binomial p-Value | ||
---|---|---|---|---|---|---|
1 | 4 | 0.007968 | 4.007968 | 0.5832 | 0.4168 | 0.5886 |
2 | 11 | 0.014024 | 11.01402 | 0.9999 | 0.0001 | 0.0006 |
3 | 3 | 0.008362 | 3.008362 | 0.3701 | 0.6299 | 0.7939 |
Critical c | Mixture Power | Rejection k | Binomial Power | ||
---|---|---|---|---|---|
0.1 | 0.1 | 2.9962 | 0.050000 | 4 | 0.012795 |
0.1 | 0.2 | 2.9962 | 0.251377 | 4 | 0.120874 |
0.1 | 0.3 | 2.9962 | 0.523352 | 4 | 0.350389 |
0.1 | 0.4 | 2.9962 | 0.757080 | 4 | 0.617719 |
0.1 | 0.5 | 2.9962 | 0.904088 | 4 | 0.828125 |
0.2 | 0.2 | 4.0086 | 0.050000 | 5 | 0.032793 |
0.2 | 0.3 | 4.0086 | 0.189362 | 5 | 0.150268 |
0.2 | 0.4 | 4.0086 | 0.415895 | 5 | 0.366897 |
0.2 | 0.5 | 4.0086 | 0.663109 | 5 | 0.623047 |
0.2 | 0.6 | 4.0086 | 0.855538 | 5 | 0.833761 |
0.3 | 0.3 | 5.0195 | 0.050000 | 6 | 0.047349 |
0.3 | 0.4 | 5.0195 | 0.171407 | 6 | 0.166239 |
0.3 | 0.5 | 5.0195 | 0.383292 | 6 | 0.376953 |
0.3 | 0.6 | 5.0195 | 0.638272 | 6 | 0.633103 |
0.3 | 0.7 | 5.0195 | 0.852383 | 6 | 0.849732 |
0.4 | 0.4 | 6.9878 | 0.050000 | 8 | 0.012295 |
0.4 | 0.5 | 6.9878 | 0.158735 | 8 | 0.054688 |
0.4 | 0.6 | 6.9878 | 0.358174 | 8 | 0.167290 |
0.4 | 0.7 | 6.9878 | 0.619691 | 8 | 0.382783 |
0.4 | 0.8 | 6.9878 | 0.856551 | 8 | 0.677800 |
0.5 | 0.5 | 7.9876 | 0.050000 | 9 | 0.010742 |
0.5 | 0.6 | 7.9876 | 0.154390 | 9 | 0.046357 |
0.5 | 0.7 | 7.9876 | 0.357879 | 9 | 0.149308 |
0.5 | 0.8 | 7.9876 | 0.645587 | 9 | 0.375810 |
0.5 | 0.9 | 7.9876 | 0.909147 | 9 | 0.736099 |
Critical c | Mixture Power | Rejection k | Binomial Power | ||
---|---|---|---|---|---|
0.1 | 0.1 | 4.0143 | 0.050000 | 5 | 0.043174 |
0.1 | 0.2 | 4.0143 | 0.386941 | 5 | 0.370352 |
0.1 | 0.3 | 4.0143 | 0.772408 | 5 | 0.762492 |
0.1 | 0.4 | 4.0143 | 0.951708 | 5 | 0.949048 |
0.1 | 0.5 | 4.0143 | 0.994442 | 5 | 0.994091 |
0.2 | 0.2 | 7.0045 | 0.050000 | 8 | 0.032143 |
0.2 | 0.3 | 7.0045 | 0.281501 | 8 | 0.227728 |
0.2 | 0.4 | 7.0045 | 0.638410 | 8 | 0.584107 |
0.2 | 0.5 | 7.0045 | 0.892613 | 8 | 0.868412 |
0.2 | 0.6 | 7.0045 | 0.983738 | 8 | 0.978971 |
0.3 | 0.3 | 9.0186 | 0.050000 | 10 | 0.047962 |
0.3 | 0.4 | 9.0186 | 0.249643 | 10 | 0.244663 |
0.3 | 0.5 | 9.0186 | 0.593093 | 10 | 0.588099 |
0.3 | 0.6 | 9.0186 | 0.874692 | 10 | 0.872479 |
0.3 | 0.7 | 9.0186 | 0.983230 | 10 | 0.982855 |
0.4 | 0.4 | 11.9910 | 0.050000 | 13 | 0.021029 |
0.4 | 0.5 | 11.9910 | 0.229635 | 13 | 0.131588 |
0.4 | 0.6 | 11.9910 | 0.562559 | 13 | 0.415893 |
0.4 | 0.7 | 11.9910 | 0.865636 | 13 | 0.772272 |
0.4 | 0.8 | 11.9910 | 0.985944 | 13 | 0.967857 |
0.5 | 0.5 | 13.9918 | 0.050000 | 15 | 0.020695 |
0.5 | 0.6 | 13.9918 | 0.224232 | 15 | 0.125599 |
0.5 | 0.7 | 13.9918 | 0.568302 | 15 | 0.416371 |
0.5 | 0.8 | 13.9918 | 0.890702 | 15 | 0.804208 |
0.5 | 0.9 | 13.9918 | 0.995777 | 15 | 0.988747 |
n | Type I Error | Power | EN0 | P (Early Stop) | Method | ||||
---|---|---|---|---|---|---|---|---|---|
0.3 | 25 | 15 | 1 | 5 | 0.033 | 0.802 | 19.5 | 0.549 | Minimax |
0.3 | 26 | 12 | 1 | 5 | 0.036 | 0.805 | 16.8 | 0.659 | |
0.3 | 27 | 11 | 1 | 5 | 0.040 | 0.806 | 15.8 | 0.697 | |
0.3 | 29 | 10 | 1 | 5 | 0.047 | 0.805 | 15.0 | 0.736 | Optimal |
0.4 | 13 | 8 | 1 | 3 | 0.031 | 0.802 | 8.9 | 0.813 | Minimax |
0.4 | 15 | 4 | 0 | 3 | 0.043 | 0.818 | 7.8 | 0.656 | Optimal |
0.5 | 8 | 4 | 0 | 2 | 0.036 | 0.836 | 5.4 | 0.656 | Minimax |
0.5 | 9 | 3 | 0 | 2 | 0.041 | 0.828 | 4.6 | 0.729 | Optimal |
0.6 | 6 | 3 | 0 | 2 | 0.015 | 0.807 | 3.8 | 0.729 | Minimax |
0.6 | 8 | 2 | 0 | 2 | 0.025 | 0.819 | 3.1 | 0.810 | Optimal |
n | Power | ESN | P (Early Stop) | |||||
---|---|---|---|---|---|---|---|---|
0.3 | 23 | 16 | 7 | 0.34 | 0.810 | 18.4 | 0.055 | 0.66 |
0.3 | 23 | 17 | 6 | 0.32 | 0.807 | 18.9 | 0.056 | 0.68 |
0.3 | 23 | 17 | 6 | 0.30 | 0.806 | 18.8 | 0.057 | 0.70 |
0.3 | 23 | 15 | 8 | 0.31 | 0.806 | 17.5 | 0.056 | 0.69 |
0.3 | 23 | 14 | 9 | 0.37 | 0.806 | 17.3 | 0.054 | 0.63 |
0.3 | 23 | 14 | 9 | 0.45 | 0.806 | 18.0 | 0.052 | 0.55 |
0.3 | 23 | 14 | 9 | 0.46 | 0.806 | 18.1 | 0.052 | 0.54 |
0.3 | 23 | 16 | 7 | 0.69 | 0.806 | 20.8 | 0.050 | 0.31 |
0.3 | 23 | 16 | 7 | 0.36 | 0.805 | 18.5 | 0.054 | 0.64 |
0.3 | 23 | 15 | 8 | 0.37 | 0.805 | 18.0 | 0.054 | 0.63 |
0.3 | 23 | 13 | 10 | 0.50 | 0.805 | 18.0 | 0.051 | 0.50 |
0.3 | 23 | 13 | 10 | 0.53 | 0.805 | 18.3 | 0.051 | 0.47 |
0.3 | 23 | 12 | 11 | 0.70 | 0.805 | 19.7 | 0.050 | 0.30 |
0.4 | 11 | 8 | 3 | 0.20 | 0.801 | 8.6 | 0.066 | 0.80 |
0.4 | 11 | 9 | 2 | 0.21 | 0.801 | 9.4 | 0.064 | 0.79 |
0.5 | 7 | 5 | 2 | 0.20 | 0.809 | 5.4 | 0.066 | 0.80 |
0.5 | 7 | 5 | 2 | 0.25 | 0.805 | 5.5 | 0.060 | 0.75 |
0.5 | 7 | 5 | 2 | 0.30 | 0.801 | 5.6 | 0.057 | 0.70 |
0.6 | 5 | 3 | 2 | 0.29 | 0.810 | 3.6 | 0.057 | 0.71 |
0.6 | 5 | 3 | 2 | 0.38 | 0.804 | 3.8 | 0.053 | 0.62 |
0.6 | 5 | 3 | 2 | 0.40 | 0.801 | 3.8 | 0.053 | 0.60 |
0.6 | 5 | 3 | 2 | 0.52 | 0.801 | 4.0 | 0.051 | 0.48 |
n | Power | ESN | ||||
---|---|---|---|---|---|---|
35 | 27 | 8 | 0.23 | 0.803 | 28.8 | 0.062 |
36 | 24 | 12 | 0.24 | 0.802 | 26.9 | 0.061 |
37 | 18 | 19 | 0.56 | 0.801 | 28.6 | 0.051 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2025 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Hutson, A.D. Optimizing One-Sample Tests for Proportions in Single- and Two-Stage Oncology Trials. Cancers 2025, 17, 2570. https://doi.org/10.3390/cancers17152570
Hutson AD. Optimizing One-Sample Tests for Proportions in Single- and Two-Stage Oncology Trials. Cancers. 2025; 17(15):2570. https://doi.org/10.3390/cancers17152570
Chicago/Turabian StyleHutson, Alan David. 2025. "Optimizing One-Sample Tests for Proportions in Single- and Two-Stage Oncology Trials" Cancers 17, no. 15: 2570. https://doi.org/10.3390/cancers17152570
APA StyleHutson, A. D. (2025). Optimizing One-Sample Tests for Proportions in Single- and Two-Stage Oncology Trials. Cancers, 17(15), 2570. https://doi.org/10.3390/cancers17152570