The Essential but Often Complex Role of Numbers and Measurements in Oncology, Continued

In the second part of this commentary on the essential but often complex role associated with interpreting numbers and measurements in oncology, I will carefully consider the clinical meaning of certain numerical statements. In doing so, I hope to emphasize the critical point that blind adherence to a number should not be permitted to negatively affect optimal care.

The example provided will again focus on ovarian cancer, due to my clinical and research interests, but similar situations can be found in multiple settings.

In the design of studies in platinum-resistant ovarian cancer, it is common to limit the number of prior regimens a patient may have received prior to entry into a clinical trial. Is there perhaps a sound principle that can be provided to justify the scientific validity of the number selected? The answer to this question is almost certainly no. In fact, many patients with ovarian cancer whose cancers are considered “platinum-resistant” may have received several platinum-based regimens before being declared to have a “resistant” malignancy if they had earlier in the natural history of their disease experienced recurrence a year or longer after the completion of primary platinum-based chemotherapy.

Do we know whether an individual with cancer who had previously responded to several platinum-based regimens is any less or more likely to experience a response to the investigative agent being evaluated in the resistant setting than a patient whose disease had quickly recurred after exhibiting modest cancer regression after a single platinum program or who had unfortunately progressed after one dose of this chemotherapeutic agent? Again, the answer to this important question is likely to be no.

However, in contrast to the issue of efficacy, it is far more likely that there will be differences between the patients described above regarding potential concerns for excessive toxicity. This is due to the effect of multiple prior regimens on normal organ function, including relevant bone marrow reserve and the risk of symptomatic neurotoxicity.

As I highlighted in the prior commentary regarding the use of a “6-month” platinum-free interval to define study eligibility or population stratification, a restriction in the number of previously received regimens is often employed to create as clinically homogeneous patient groupings as possible in which to examine both efficacy and toxicity. Of course, the creation of reasonably comparable research populations is most relevant in a phase 3 randomized trial designed to compare study groups.

Therefore, the trial-inclusion restrictions described above are appropriate; for example, patients who have received more than 2 prior therapeutic regimens being declared ineligible for study inclusion.¹

However, it is critical to acknowledge that this valid study-based decision does not, by itself, represent either a scientifically or clinically meaningful justification for denying the potential benefit of a given strategy to a patient who has received more than this number of regimens outside the investigative setting.

It is not difficult to find other examples where the goal associated with the use of numbers and measurements in oncology is understood and even laudable, but where their application is problematic. Several examples will be helpful.

Standard criteria for admission to many hospice programs require a physician to certify that a patient is anticipated to live for less than 6 months from the time of entry.² Again, the reasonable intent here is to provide end-of-life care benefits for a limited but still clinically meaningful period.

Unfortunately, although prediction of very short survival can be made with reasonable medical certainty, such as in the presence of rapidly declining cardiac function, it is far more difficult to predict the clinical course of patients with advanced cancers. This is especially the case when the goal is to provide essential care during the final months of a patient’s life, rather than only the closing days.

A second example is the common and appropriate request of patients with cancer and their families to inquire about the possible anticipated prognosis for survival. Oncologists understand the importance of emphasizing in this discussion the statistical nature of the data and the fact that it is usually not possible at the beginning of an individual’s cancer journey to know where the patient will fall within the observed continuum.

The concern being highlighted here is not with the need to present analytical tumor stage data (1-year, 5-year survival rates, etc), but rather the serious limitation that in most clinical settings, the only currently available outcome information is specific to the time of initial diagnosis. Therefore, the question to be asked is as follows: What does an oncologist tell a patient who has successfully completed all therapy for a particular stage III cancer and is now 2-plus years from diagnosis, regarding the statistical likelihood of their being alive at 5 years (or longer) from the date of diagnosis?

Fortunately, investigators have begun to publish data in response to such inquiries (conditional survival analysis).³ However, this is an area that requires expanded effort to make the critically relevant measurements of statistically anticipated survival following increasingly beneficial therapy as current, objective, and clinically meaningful as possible.

Finally, we conclude with my “favorite measurement”––the size of the largest residual tumor mass remaining within the peritoneal cavity at the completion of primary cytoreductive surgery for stage III or IV ovarian cancer.⁴ While the goal of such surgery is to remove all gross (visible) cancer, the standard criteria for “small volume residual disease” is no mass above 1 cm in maximum diameter. Is it realistic to believe an experienced and busy gynecologic cancer surgeon will spend the required time measuring all remaining visible unresectable masses within the abdominal cavity (which may be considerable in number) to determine that none are physically measured to be larger than 1 cm in maximum diameter?

The folly of this suggestion was highlighted to me many years ago in my role auditing surgical records for the inclusion of appropriate protocols in Gynecologic Oncology Group phase 2 studies examining intraperitoneal chemotherapy. The maximum single residual mass size for patient eligibility for several studies was 1 cm. More than one surgeon, apparently recognizing the effort required, desiring their patients to be permitted study entry, and perhaps appreciating this somewhat absurd request to measure mass size, recorded “0.99 cm” as the maximum tumor diameter.

Editor’s note: To read the first part of this commentary, click here.

References

1. Pujade-Lauraine E, Hilpert F, Weber B, et al. Bevacizumab combined with chemotherapy for platinum-resistant recurrent ovarian cancer: the AURELIA open-label randomized phase III trial. J Clin Oncol. 2014;32(13):1302-1308. doi:10.1200/JCO.2013.51.4489
2. Wagner V, Marks A. What is hospice? JAMA. 2024;332(14):1218. doi:10.1001/jama.2024.13144
3. Hieke S, Kleber M, König C, Engelhardt M, Schumacher M. Conditional survival: a useful concept to provide information on how prognosis evolves over time. Clin Cancer Res. 2015;21(7):1530-1536. doi:10.1158/1078-0432.CCR-14-2154
4. Chase DM, Mahajan A, Scott DA, Hawkins N, Kalilani L. The impact of varying levels of residual disease following cytoreductive surgery on survival outcomes in patients with ovarian cancer: a meta-analysis. BMC Womens Health. 2024;24(1):179. doi:10.1186/s12905-024-02977-5