The term “ketogenic diet” (KD) generally describes any diet under which the serum levels of the ketone bodies β-hydroxybutyrate (BHB), acetoacetate (AcAc) and acetone increase beyond their reference ranges, typically 0.3 mmol/l for BHB, the most abundant circulating ketone body. This state of “ketosis” is a physiological condition for humans also occurring naturally in neonates, during intermittent and longer-term fasting or after intense exercise without glucose replacement [1]. To achieve the state of “nutritional ketosis”, a KD is usually composed of a very low carbohydrate (CHO) and high fat content and may additionally be combined with calorie restriction.

In recent years, the number of expert reviews concluding that KDs possess promising effects for cancer patients has sharply increased [2–10]. Specifically, it has been proposed that ketone bodies would confer protection to normal cells during radio- or chemotherapy while being non-protective or even toxic to tumor cells [6]. In addition, a high fat diet with ample protein intake may be ideally suited to meet the metabolic demands of the cancer patient whose metabolism is often affected by the tumor in such a way that peripheral tissues become more CHO “intolerant” while relying more on fatty acid oxidation [10]. However, clinical research on the effects of KDs in cancer patients is still scarce. There are some concerns against using KDs in cancer patients due to the side-effects that can occur. These include deficiencies in micronutrients, appetite loss, nausea, constipation, fatigue, hyperlipidemia and—especially relevant for cancer patients weight loss [11]. Some of these side-effects, however, may be attributable to the transition phase in which the body has to adapt to the high fat and low CHO content of the diet, usually one to 3 weeks. Other, longer-term side-effects may be preventable by restricting the KD to the timeframe of (RT)/radiochemotherapy (RCT) since there is evidence that it is most useful as a supportive intervention in this context [6]. We have therefore started to collect data on ambulatory patients consuming a KD during RT, and here report our initial clinical experience with the first six of such patients within a busy community hospital environment. Our main aim was to assess the feasibility of such an intervention and track body composition changes during the combined KD-RT treatment; secondary outcome variables included biochemical blood parameters and quality of life (QoL). While these data are unable to provide information about putative anti-tumoral effects of the diet, they might add further evidence to the hypothesis that KDs in cancer patients are safe and feasible as supportive measures during RT.

Prior to the initiation of this study we received approval by the institutional ethics review board. Data collection was performed on the first six consecutive patients undertaking a KD during RT/RCT in our clinic. In most cases patients already had initiated or wished to initiate the diet at the initial consultation with the treating radiation oncologist. In one case (patient 2) we recommended the diet during RCT in an attempt to stop continuing weight loss. All patients were told, however, of the limited clinical experience with this diet in the context of cancer treatment and informed about potential side effects such as headaches, light-headedness, constipation/diarrhea and weight loss. Exclusion criteria were defined a priori as Karnofsky index less than 80, type I diabetes, pregnancy, pace maker, cognitive impairments or rare metabolic diseases that would contradict a KD such as defects in key enzymes of ketogenesis, ketolysis, gluconeogenesis or fatty acid oxidation. At study entry, all patients gave written informed consent to collect and publish their data. In particular, all patients agreed that their data could be used for scientific purposes such as publication in a scientific journal. Besides the KD and QoL assessment none of the interventions were unusual for a patient undergoing RCT in our clinic. All these procedures were performed according to the ethical standards of the declaration of Helsinki and approved by an institutional review board.

The results of the body composition analysis are given in Table 4 and displayed in Figs. 1 and 2. Changes in patient 5 were only computed for the 73 days that this patient followed a KD (see below for more details). Weight loss occurred in all patients during the combined RT/KD, but changes were only significant in patients 3 and 5. Unfortunately, patients 2 and 4 had metallic parts within their bodies that made body composition estimates unreliable. BIA estimates in the remaining four patients suggested that body mass loss was mainly composed of FM which decreased significantly in patients 3, 5 and 6 in absolute as well as relative terms. In contrast, FFM did not change significantly in absolute terms but increased significantly relative to body weight in patients 3, 5 and 6 (Table 4). No significant changes in ECW, TBW or hydration status (ECW/ICW) where observed, but ICW decreased significantly in patients 1, 3 and 5 and increased significantly in patient 6. No significant changes in PA occurred with the exception of a decrease in patient 5 (Fig. 2).

Slope β	Patient 1	Patient 2	Patient 3	Patient 4	Patient 5 a	Patient 6
BW (kg/week)	−0.58 [−1.35, 0.18]	−0.21 [−1.59, 1.17]	−0.71 [−0.90, −0.53]	−0.06 [−2.29, 2.18]	−0.84 [−1.07,−0.61]	−0.22 [−0.43,0.01]
Relative BW (%/week)	−0.64 [−1.45,0.21]	−0.31 [−2.25,1.65]	−0.93 [−1.18,−0.68]	−0.07 [−2.33,2.25]	−0.74 [−0.94,−0.54]	−0.38 [−0.78,0.01]
PA (°/week)	−0.02 [−0.10,0.05]	−0.02 [−0.16,0.12]	−0.02 [−0.07, 0.03]	−0.14 [−1.8, 1.6]	−0.06 [−0.08,−0.04]	0.06 [−0.002,0.12]
FM (kg/week)	−0.40 [−1.02,0.22]	NA	−0.49 [−0.68,−0.30]	NA	−0.68 [−0.90,−0.45]	−0.28 [−0.48,−0.09]
Relative FM (%/week)	−0.21 [−0.63, 0.23]	NA	−0.49 [−0.75, −0.23]	NA	−0.29 [−0.54,−0.04]	−0.42 [−0.76, −0.10]
FFM (kg/week)	−0.13 [−0.61,0.35]	NA	−0.22 [−0.45,0.02]	NA	−0.17 [−0.53,0.21]	0.07 [−0.18,0.32]
Relative FFM (%/week)	0.23 [−0.14,0.61]	NA	0.49 [0.37,1.46]	NA	0.29 [0.04,0.54]	0.42 [0.09,0.75]
ECW (l/week)	−0.023 [−0.31, 0.26]	NA	−0.063 [−0.18, 0.054	NA	−0.004 [−0.18,0.18]	−0.069 [−0.28, 0.14]
ICW (l/week)	−0.15 [−0.30, −0.01]	NA	−0.15 [−0.27, −0.06]	NA	−0.13 [−0.25, −0.008]	0.18 [0.04, 0.32]
Hydration (ECW/ICW) (%/week)	0.46 [−0.49, 1.42]	NA	0.16 [−0.22, 0.53]	NA	0.39 [−0.15,0.91]	−1.23 [−2.65, 0.19]

Trends are considered significant if the 95 % HPD interval excludes 0, with positive slopes indicating an increase, negative slopes a decrease in the quantity under consideration

^a Trends in patient 5 where only computed for the 73 days he followed a KD

Below, a brief description of interesting results for some individual patients is given.

In this paper we describe our initial experience with the first six patients who undertook a KD during RT. While the small sample size, short duration of follow-up and lack of a control group are obvious limitations for the interpretation of our results the strength of our data is that patients were followed prospectively and many measurements of clinical relevance were made. A main result is that the KD intervention within the context of a busy community hospital setting was feasible, but posed some challenges, the main one being patient compliance which is hard to control within an ambulatory setting. From our experience with the six cases presented here we have learned that three measures would help to maximize compliance and high ketosis: (1) frequent individual counselling by a dietician experienced with KDs; (2) provision of ketogenic formulas and meals; (3) offering cooking classes. These three steps have now been implemented in our clinic for the care of future patients who wish to undertake a KD during RT.

Importantly, no diet-related side effects occurred in our patients who finished RT without breaks and rated their subjective feeling on the diet as “good”. Furthermore no negative effects on blood parameters were detectable, contradicting the account of some authors who describe the KD as unhealthy or even dangerous. It is likely that limiting the duration of a KD to a short timeframe such as the period of RT limits any negative impact on blood parameters. Weight loss could be a concern, however, although in our study this has to be interpreted within the context of RCT that by itself can lead to BW loss of a similar magnitude as observed in our patients [19]. While theoretically a KD has both appetite stimulating and appetite suppressing mechanisms, the interaction of these mechanisms as well as the common experience in humans speaks more for an overall anorexigenic effect [20], consistent with the reports of our patients.

Our BIA data indicate that weight loss was mainly due to FM loss during the diet phase. PA was also basically unaltered with the exceptions of patient 5 who had extensive disease with a rapidly growing tumor and the steep decrease of PA at the final measurement in patient 4 that remains open for interpretation. PA is an important measure of overall health, muscularity and nutritional status on the cellular level [21] and it is not uncommon for PA to decrease during RT [19].

While a decrease in FM can even be rated as beneficial for some patients [22] there is strong evidence that skeletal muscle mass is the crucial component for long-term prognosis and QoL [23]. It would be ideal for a diet of cancer patients to both counteract protein catabolism and stimulate protein anabolism. The KD might be such a “cancer-specific diet” [10]. Regarding catabolism, studies have shown that ketosis decreases urinary nitrogen losses and muscle mass breakdown in both healthy individuals [24] and undernourished cancer patients [25]. Regarding anabolism it should be considered that different protein sources differ in their net nitrogen utilization (NNU), i.e., the percentage of amino acids that—after digestion—get utilized in the anabolic pathway to build new body protein without energy production or nitrogen loss [26]. Proteins with lower NNU provide a larger percentage of amino acids that get deaminated in the catabolic pathway, delivering more precursors for endogenous glucose production via gluconeogenesis as well as nitrogen catabolites that have to be detoxified by the liver. For this reason we advised our patients to consume high quality proteins such as fish and eggs and began substituting them with MAP which has been shown to have a NNU of 99 % when used as the sole protein source in healthy individuals [26].

Focussing on high quality protein makes particular sense in the context of a KD where the ketosis-repressing properties of dietary protein have long been known. The classic KD used for treating intractable epilepsy in children has therefore been based on calorie restriction and a ketogenic ratio of at least 3:1 in order to maximize ketosis and its therapeutic efficacy [27], with the downside of severe protein restriction. The ketogenic ratio in our patients ranged from approximately 1:1–2:1 (but probably higher due to not reporting of cooking fats and dressings) which allowed for ample protein intake and good tolerability, but might have been too low to considerably elevate BHB concentrations in some patients. Our experiments with patient 5 suggest that ketogenic drinks, especially enriched with medium chain triglycerides, provide an easy way to transiently increase the ketone to blood glucose ratio.

Given the large variety of human tumors and the fact that most cancer patients are old and/or present with weight loss, sarcopenia, diabetes and other metabolic comorbidities (including those induced by their tumor) it is presently unclear what the ideal ketogenic ratio for an individual cancer patient is. Thomas Seyfried’s group has emphasized the importance of maximizing the ketone to blood glucose ratio [28]. Can we achieve a good compromise between maximally lowering this ratio and providing sufficiently high protein and energy intake in order to maintain important muscle tissue? Can we also implement additional calorie restriction in certain patients to enhance efficacy without a negative impact on body composition as indicated by one case report [29]? These questions must be addressed with human data. For our part, the initial observations reported here motivated us to initiate a randomized study on the feasibility and effects of a KD intervention on body composition, blood parameters and QoL. The protocol of this study called KETOCOMP (ClinicalTrials.gov identifier: NCT02516501) is provided in a companion article [30].

We presented a variety of prospectively collected measurements on the first six consecutive patients that undertook a KD during RT treatment in our clinic within a busy community hospital. No serious diet-related side effects occurred, and subjective feeling on the diet was rated as good. Ketosis was not as pronounced as it is usually reported for protein- or calorically restricted KDs, and blood sugar levels did not decrease significantly. This might indicate problems with either compliance or prescribed diet composition which is discussed within the context of the ideal diet composition for cancer patients aiming at the preservation of muscle mass. Importantly, although significant weight loss occurred in some patients, our BIA data indicate that weight loss was mainly due to fat mass loss. From these observations we implemented additional measures of diet counselling and initiated a randomized study (ClinicalTrials.gov identifier: NCT02516501) the protocol of which is published in a companion paper [30].