When Toxicity Counts: It's Time to Reappraise Non-Severe but Clinically Relevant Side Effects

Maurie Markman, MD

As an increasing number of cancers are considered chronic diseases treatable with continuous (ie, maintenance) drug delivery, it is essential that our vocabulary describing the side effects experienced by our patients more appropriately reflects this new paradigm.

Our concepts of measuring the toxic side effects associated with antineoplastic drug therapy were fashioned in an era where such treatment was generally relatively limited in duration ( <6 months), where serious and even life-threatening toxicities were the greatest concern (eg, grade 4 neutropenia, thrombocytopenia, mucositis), and where cytotoxic therapy was most commonly delivered by the intravenous route. In this earlier era, patients generally received a single che-motherapy regimen for metastatic cancer, or at most, perhaps two regimens. Survival for patients presenting with stage IV cancers, or where the tumor recurred after initial local therapy, was also quite limited, and in the majority of situations was less than one year or a maximum of two years in duration.

The standard drug development paradigm during this time included the conduct of one or more phase I trials designed specifically to define the maximally tolerated dose to be utilized in future phase II and phase III studies, and likely ultimately in clinical practice, assuming an acceptable degree of efficacy was subsequently documented. This approach was employed for both novel agents introduced into the clinic and unique combinations of routinely utilized antineoplastic drugs.

In many trials, the phase I dose and schedule deemed acceptable actually was frequently determined based solely on what could be reasonably well tolerated during a single treatment cycle, or at most, two cycles. This occurred principally because in typical phase I trials, objective responses were rarely observed and progression was often quite rapid. Further, chronic or prolonged use of the strategy was rarely, if ever, contemplated.

Finally, and most relevant to the specific point of this commentary, the dose-limiting toxicity in these regimens was in the vast majority of circumstances defined by the incidence of severe and relatively easily measurable side effects, such as two of three or three of six patients demonstrating grade 3 or 4 bone marrow suppression. This clinical trial paradigm was mandated regardless of whether the sponsor of a trial was a pharmaceutical company, an individual academic investigator, or a not-for-profit multi-institution cooperative group.

Table 1. Gastrointestinal Disorders, Adverse Event Criteria

Nausea

Definition

A disorder characterized by a queasy sensation and/or the urge to vomit

Grade 1

Loss of appetite without alteration in eating habits

Grade 2

Oral intake decreased without significant weight loss, dehydration, or malnutrition

Grade 3

Inadequate oral caloric or fluid intake; tube feeding, total parenteral nutrition, or hospitalization indicated

Cancer Therapy Evaluation Program. Common Terminology Criteria for Adverse Events (CTCAE) Version 4.03. National Cancer Institute website. http://evs.nci.nih.gov/ftp1/CTCAE/CTCAE_4.03_2010-06-14_ QuickReference_8.5x11.pdf. Published May 28, 2009. Updated June 14, 2010. Accessed September 27, 2012.

New Regimens Change Dynamic

But a new era is upon us. Oral antineoplastic drug administration is increasingly common, and it is not unusual for responding patients to remain on a treatment regimen for prolonged periods of time (eg, greater than one or even many years). Further, in several settings involving a number of tumor types, “maintenance therapy” following the attainment of a maximal response (complete, partial, or stable disease) is currently the standard of care, or at least commonly considered a reasonable approach in routine disease management.

In this scenario, it can be argued that lower-grade intermittent, or continuous and far less quantifiable, side effects may become extremely distressing and seriously impact an individual patient’s quality of life. Further, as a consequence of such outcomes, an individual patient might even abandon quite effective therapy.

Consider, for example, the daily occurrence of so-called non-severe grade 1 or grade 2 nausea experienced by a patient taking an antineoplastic pill on a regimen that is anticipated to last for 18 months or longer (Table 1). Is this realistically a tolerable regimen? Yet, such an outcome would likely not be considered a serious adverse event in many cancer trials despite the potential negative impact on an individual’s quality of life or the patient’s ability—and willingness&mdash;to continue treatment.

Two Examples Point to Problems

It is fair to inquire whether this is really a relevant issue in clinical trials, or merely an unlikely theoretical concern. While simply only one illustration of the issue, an appropriate response to the question is provided by the results of a recently reported placebo-controlled, randomized phase II trial evaluating tasquinimod, a novel orally administered (2 times/day) antineoplastic agent in men with minimally symptomatic metastatic and castration-resistant prostate cancer.¹

While a greater incidence of grade 3 and 4 adverse events were observed in the patients receiving active treatment, perhaps the most striking feature of the toxicity profile noted in this trial was the proportion of individuals who discontinued treatment due to toxicity: 22% in the tasquinimod arm versus 1% in the placebo arm. Further, withdrawal from therapy frequently occurred before any protocol-defined dose modification. In fact, while only 1% of patients receiving active therapy reported grade 3-4 nausea, one-quarter of the population noted nausea that was assigned as being a protocol-defined grade 1 or 2 adverse event (Figure).

Figure. Patients’ Experiences in Tasquinimod Trial

Although the toxicity profile of tasquinimod was deemed acceptable in this randomized study, 22% of the participants in the treatment arm, top chart, withdrew because of adverse events, most of which were grades 1 and 2, bottom chart.

J Clin Oncol

Adapted from Pili R, Häggman M, Stadler WM, et al. Phase II randomized, double-blind, placebo-controlled study of tasquinimod in men with minimally symptomatic metastatic castrate-resistant prostate cancer [published online ahead of print September 19, 2011]. . 2011;29(30):4022-4028. doi: 10.1200/JCO.2011.35.6295.

However, imagine the quality of life of a patient taking an oral antineoplastic agent twice a day who experiences grade 1 or 2 nausea every day after taking the medication. Is it any wonder that nearly one in four men receiving active therapy elected to withdraw from the treatment program?

Likewise, consider the prolonged administration of a platinum agent or, more likely, a taxane-based treatment regimen in a patient who has achieved an excellent response to primary chemotherapy but where the individual notes grade 1 peripheral neuropathy, which is considered mild numbness or tingling of the hands and/or feet that does not interfere with normal function. What exactly does this specific symptom really mean to the quality of life of an individual patient if the side effect is constantly present or persists for days or even weeks after each course of therapy?

For example, in a phase III randomized trial that explored the continuation of 12 months of single-agent paclitaxel in women with advanced ovarian cancer who had attained a clinical complete response to primary platinum-paclitaxel chemotherapy, only 5% of patients treated with this strategy were reported to have experienced severe (grade 3) neuropathy, but grade 2 effects were noted in an additional 18% of treated individuals (Table 2).² Further, it is reasonable to speculate that, if carefully investigated and subsequently recorded, grade 1 neurological discomfort/pain would have been an issue in an even greater percentage of the treated population.

Table 2. Adverse Events Among Patients in Paclitaxel Trial

J Clin Oncol

Adapted from Markman M, Liu PY, Wilczynski S, et al. Phase III randomized trial of 12 versus 3 months of maintenance paclitaxel in patients with advanced ovarian cancer after complete response to platinum and paclitaxel-based chemotherapy: a Southwest Oncology Group and Gynecologic Group trial. . 2003;21(13):2460-2465.

Definitions Falling Short

Does the presence of a low-grade neuropathy, characterized as grade 1, interfere with a patient’s activities of daily living (eg, the ability to walk or sleep due to “mild” pain/discomfort), the enjoyment of long-established hobbies (eg, playing a musical instrument), or even suggest a potential noncancer—related safety concern (eg, impact on reflexes when driving an automobile)? Do data generated in early-phase trials, or even later-stage registration studies, truly capture the impact of such low-grade but persistent side effects on the life of an individual patient?

And, in the absence of such information, how many patients treated outside the confines of a clinical trial will abandon documented effective antineoplastic therapies due to the development of “low-grade” but clearly subjectively unacceptable side effects of such treatment?

Unfortunately, without such population-based data, it is most unlikely that the clinical oncology research community will even know if it is necessary to study this issue, and perhaps work to find an alternative or modified therapeutic approach to maximize clinical utility while minimizing the risk of such low-grade toxicity.

Other low-grade events worthy of further clinical investigation into their overall and treatment-specific impact on a patient’s quality of life and treatment discontinuation include gastrointestinal effects (eg, mucositis, stomatitis, diarrhea), skin rash, anemia, and fatigue. Again, imagine the impact of “grade 1 toxicity” associated with any of these events that are potentially present for one, two, three, or even more years.

Patient-Focused Outlook Needed

As an increasing number of cancers enter the realm of being considered as chronic disease processes, where prolonged survival is observed in association with continuous (ie, maintenance) drug delivery, it is essential that our vocabulary describing the side effects experienced by our patients more appropriately reflects this new paradigm.

Further, for prolonged continuous systemic antineoplastic drug delivery realistically to be a viable management option, it is essential that future research focus on the development of strategies—either the drugs themselves or the associated co-administration of supportive care medications&mdash;that minimize these “low-grade” but potentially truly distressing side effects.

Examples of such approaches might include employing lower drug dosages during maintenance therapy, utilizing a modified administration schedule (eg, every other day, rather than daily delivery), or treating patients for a period of time, followed by a short “drug-free holiday.”

Ultimately, well-designed and conducted clinical trials will be required to define the optimal strategies to be utilized in these settings, but such efforts will only be undertaken if the oncology community truly recognizes both the importance and the potential seriousness of the issues involved.

Maurie Markman, MD, editor-in-chief, is senior vice president for Clinical Affairs and national director for Medical Oncology at Cancer Treatment Centers of America. maurie.markman@ctca-hope.com

REFERENCES

1. Pili R, Häggman M, Stadler WM, et al. Phase II randomized, double-blind, placebo-controlled study of tasquinimod in men with minimally symptomatic metastatic castrate-resistant prostate cancer [published online ahead of print September 19, 2011]. J Clin Oncol. 2011;29(30):4022- 4028. doi: 10.1200/JCO.2011.35.6295.
2. Markman M, Liu PY, Wilczynski S, et al. Phase III randomized trial of 12 versus 3 months of maintenance paclitaxel in patients with advanced ovarian cancer after complete response to platinum and paclitaxel-based chemotherapy: a Southwest Oncology Group and Gynecologic Group trial. J Clin Oncol. 2003;21(13):2460-2465.